Stakeholders’ perceptions on earnings management motivations and techniques in Libyan commercial banks

The issue of earnings management has continued to be problematic in the financial reporting context. It has proved to be an important topic that concerns a wide range of stakeholders including regulators, investors and preparers. This paper examines the perceptions of Libyan Commercial Banks' (LCBs) stakeholders regarding earnings management (EM) motivations and techniques. The paper reports on a questionnaire survey of stakeholders which yielded 102 Responses (response rate 53%). It is found that LCBs’ managers are engaged in EM practices as a result of a number of motivations. In consistent with accounting literature on EM, LCBs’ managers tend to use Loan Loss Provision LLP as well as other techniques to manage their earnings. However, the research finding reveal some other “techniques” that are being used for EM. Some of them clearly lay out of GAAP and law framework which may be seen as an explicit breach to bank managers’ accountability. The importance of the banking sector for society has been amply demonstrated in recent years. The evidence provided by the questionnaire results refer to a serious problem to the accountability of LCBs’ managers. The existence of these motivations, which appear to be unavoidable, put pressure on the accountability process and expose it to a lack of trust and disrepute which therefore may have an adverse effect on the relationship between LCBs’ managers and their stakeholders. LCBs’ manager should be free from such motivations in order to produce unbiased and fair accounting information.


Stakeholders' perceptions on Earnings management motivations and techniques in Libyan
Commercial Banks

Introduction
Literature on earnings management attributes the existence of such a practice due to conflict of interest between owners and managers (Wu et al., 2016). This managerial behaviour, according to Aerts et al. (2013) is mainly incurred for "the benefit of insiders" by acting as to "mislead outsiders' perceptions" about the firm's financial performance (p. 94). Incentives represent the rational basis for earnings management practice; without those incentives, managers should not be deterred from choosing accounting judgments and methods that fairly represent the economic performance of the firm. In many cases, managers may find themselves in situations in which they are tempted to manipulate the reported profits because the firm's value and mangers' wealth are associated with the reported earnings (Jackson and Pitman, 2001). Despite the fact that earnings management is widespread (Levitt, 1998), it is difficult for researchers to document it with credible proof as verifying whether earnings have been managed is not an easy task. Identifying the manger's incentives to manage earnings and estimating whether patterns of otherwise unexpected accruals are consistent with these incentives could help (Healy and Wahlen, 1999).
Financial reporting quality research on developed countries has extensively been reported by literature. However, very little attention was given to emerging markets including like Libya where there are calls for more research to investigate this issue (Zakari and Menacere, 2012;Sawan and Alsaqqa, 2013;Sawan and Alzeban, 2015;Barghathi etl al, 2017). This paper explores the Libyan Commercial Banks's stakeholders' perceptions in relation to the motivations and managerial incentives that induce bank managers to involve in earnings management. The stakeholders' perceptions regarding earnings management techniques i.e. how earnings are manged by LCBs' manager are also investigated.
Our study contributes to the literature in two respects. Primarily, it is perhaps the first study in Libya to examine the EM motivations and techniques. Previous studies particularly on Libya have focused only on the existence of EM (Elseraiti, 2011;Barghathi et al., 2017). The present paper fills the research gap by examining why LCBs' managers are engaged in EM and how earnings can be managed. Second, we also add to the earnings management literature in general and in Libya in particular by eextending the prior work of Barghathi et al. (2017) who confirmed the existence of EM.
The remainder of this paper proceeds as follows. Section 2, provide background on literature review on earnings management motivation and techniques. Also, the section develops the main research questions. Section 3 presents the research methodology. Section 4 reports the research results while section 5 concludes the paper.

Literature review
Bank managers are much more concerned about earnings stability and growth; therefore it is anticipated that bank managers are engaged in earnings management (Bhat, 1996). Banks generally represent a significant proportion of total listed companies which means that banks have an influential role in the capital market (Kanagaretnam et al. 2010). Shen and Chih (2005) stressed the importance of banks and describe the banks' share in the capital market as "typically large".
Moreover, banks play a vital role in economic development, with investors and regulators monitoring banks' performance on a regular basis; the former for monitoring share prices, the latter to assure the robustness of a banking system's financial soundness. As a result, reported earnings growth remains one of the key pointers that demonstrate a bank's performance and financial stability which ultimately suggests that bank managers may be inclined to smooth earnings volatility over periods. By earnings management in general and income smoothing in particular, bank managers can help to sustain the appearance of a robust financial position as well as meeting legal requirements (Taktak et al., 2010a). It has been suggested that income smoothing is a continuing practice that is employed by banks (Bhat, 1996). Moreover bank managers are accused of being more likely to indulge in earnings manipulations compared to others (Leventis, 2011).
For instance, Bhat (1996) suggests that banks' managers exercise income smoothing for a number of reasons: (a) to enhance the risk perceptions of the bank to its investors and regulators; (b) to support managers' efforts in maintaining their compensation schemes; (c) to satisfy shareholders where income smoothing will enable managers to afford a constant stream of dividends; and (d) to provide low quality managers with a good chance of delivering an image of high quality management to investors where measuring management's quality is difficult. Also, banks' managers may smooth earnings to reduce tax payments and improve share prices.
An overview of the literature related to earning management motivation and earnings management techniques are reported in the following two section.

Managerial Incentives for Earnings Management
Four kinds of incentives have been identified as inducing managers to be involved in earnings management: external contract incentives; management compensation contract incentives; regulatory and political motivations; and capital market motivations (Noronha et al., 2008). These motivations are briefly discussed in the following section.

External Contract Incentives
Some external contracts, for example debt contracts, dividend covenants and supplying contracts contain provisions by which a company agrees to reach a certain level of earnings, debt, or limit payments to the shareholders (Noronha et al., 2008). When the company is close to violating one of the covenants, managers may be motivated to misrepresent the accounting data and therefore the reported earnings so they can meet the contract requirements and avoid default by increasing the reported income (Noronha et al. 2008;Duncan, 2001;Jackson and Pitman, 2001). Some lending contracts are based on accounting numbers and designed in a way to restrict certain types of managers' decisions e.g. "value-reducing investment and financing decisions" which, according to Watts and Zimmerman (1986, p. 210), provide incentives to managers for accounts manipulation. They hypothesized that managers are expected to manipulate earnings when a "firm's debt/equity ratio" (p. 216) becomes relatively large. The literature reports other evidence that firms which are close to lending covenant limits are engaged in earnings management, for example Defond andJiambalvo (1994 cited in Noronha et al., 2008) in their study found that firms did accelerate earnings in the year prior to a covenant violation. Iatridis and Kadorinis (2009) found out that UK listed firms which are close to debt covenant violations employ earnings management practices in order to not default the covenants.

Management Compensation Contract Incentives
Accounting figures represent the basis to control and regulate contractual relationships among a company's stakeholders; management compensation is such a relationship (Healy and Wahlen, 1999). Companies may link the bonus programmes of managers with their economic performance in order to bring into line management's' goals with those of shareholders which in return provides a very strong incentive for managers to employ the reported earnings to increase their compensation payments (Duncan, 2001). Healy (1985) blamed management compensation for creating incentives that induce managers to engage in earnings management. Management compensation which is regarded as the "primary incentive for earnings management" (Achilles et al., 2013, p. 581) is basically placed to combine both managers' and shareholders' interests as these two groups' interests can be conflicted (Wolk et al., 2008). However, according to Watts and Zimmerman (1986), managers of firms with earnings-based bonuses are expected to manipulate the reported earnings to their own advantage. In the UK, listed firm managers, according to Iatridis and Kadorinis (2009), are improving their financial numbers through earnings management practices so that their compensation is protected or increased.
Mangers may also attempt to manipulate reported income out of concern for job security. When current earnings are low while next year's profit is expected to be relatively high, mangers may advance some earnings from the next period to the present and in contrast when the current year's profit is high and next year's earnings are expected to be low, mangers may also shift some earnings to the next year. Moreover, managers typically believe that reporting a growing steady stream of income is highly appreciated by financial analysts (Jackson and Pitman, 2001). In addition, individuals may be motivated to get a promotion based on the fact that people may be very obsessed with climbing the corporate ladder; this aspect along with a company policy to reward ambitious persons, means that seeking or gaining a promotion can be an incentive to manipulate earnings (Duncan, 2001).

Regulatory and Political Motivations
Regulatory incentives to manage earnings may be very significant in cases where reported earnings will affect the procedures of regulatory or government officials. Commercial banks in Libya, for example, are monitored by the Central Bank of Libya for compliance with regulation and in such a case, managers may be motivated to manipulate the reported earnings in order to avoid the actions that could be taken by such a governmental and regulatory body (Jackson and Pitman, 2001). Moreover, some banking regulations require banks to maintain a certain minimum level of capital adequacy requirements that are represented in the form of accounting figures. Such regulations may tempt managers to manipulate these figures to meet the requirements (Healy and Wahlen, 1999).
In addition, tough taxation laws may induce firms to manipulate their reported income. According to Watts and Zimmerman (1986), managers are motivated to manipulate the reported earnings to avoid political costs, they hypothesised that managers are more likely to manage the reported earnings downwards to lessen the tax expenses. Baralexis (2004) found out that small firms in Greece employ earnings management to gain tax savings. Likewise, Gonchanalyze and Zimmermann (2006) tested the earnings management practices of Russian companies in the years 2001 and 2002 to find out that Russian companies lessen their reported income in order to reduce income taxes.

Capital Market Motivations
Meeting revenue expectations and analysts' predictions may be the main catalysts that induce managers to manipulate earnings (Magrath and Weld, 2002). The capital market incentive stems basically from the connection between reported earnings and a company's market value, due to the fact that stock markets typically respond negatively to companies that fail to meet analysts' earnings expectations. Companies which are not reaching these predictions may be involved in earnings management to satisfy analysts' expectations (Jackson and Pitman, 2001). Eventually, failing to meet analysts' expectations can lead to adverse consequences on a company's stock price e.g. when Procter & Gamble declared that the company would not reach the analysts' forecast in the first quarter of 2000, its stock price decreased by 30% (Duncan, 2001). Literature suggests that managers are concerned about meeting or beating the analysts' expectations and that it is a "fundamental" target since stock market severely reacts to negative reported earnings (relative to expectations), moreover, the positive earnings (relative to expectations) are rewarded which, according to Athanasakou et al. (2009) gives managers a strong incentive to manipulate the reported earnings. Glaum et al. (2004), for example, compared the earnings management incentives of US and German firms; they found that both U.S. and German firms are managing their earnings to avoid losses and decreases in earnings. Evidence was reported by Athanasakou et al. (2009) who examined UK firms' engagement in earnings management for the purpose of meeting analyst earnings expectations. Their findings suggest that UK firms tend to manipulate their income to meet the analysts' expectation and to avoid reporting negative earnings. Kamel and Elbanna (2010) examined the potential incentives for engagement in earnings management in Egypt to find out that Egyptian firms are mainly engaged in earnings management for the purpose of, among others, reporting profits and avoiding reporting losses as well as achieving high-share valuation.

Earnings Management in the Banking Industry
The literature reports that banks' managers use the LLP as a mechanism for earnings manipulation (Anandarajan et al. 2003(Anandarajan et al. , 2007Kanagaretnam, 2010;Leventis et al. 2011;DeBoskey and Jiang, 2012). GAAP offers latitude in choices available to account for a specific set of financial events which gives the opportunity to smooth income through choosing from accounting alternatives.
Given the considerable scope for banks' managers when estimating the amount of LLP, banks' managers may smooth earnings by manipulating LLP (Bhat, 1996;Kanagaretnam et al., 2003).
LLPs are designed to reflect the sum of funds that are likely to be lost in the future (Bhat, 1996).
LLPs reveal the expected future loan losses to be disclosed in the current period as accrued expenses on the income statement (Whalen, 1994). However, "federal banks and securities regulators" realize that the LLPs that have been anticipated by bank managers cannot precisely match real losses and thus include a margin for inaccuracy, a margin that has been exploited by bank managers (Anandarajan et al. 2007). According to Whalen (1994), bank managers possess private information about the default risk inherent in LLPs and as a result their judgement in estimating LLPs each period is essential. He added that investors and regulators cannot obtain all of bank managers' information as it is "prohibitively costly" and accordingly "bank managers can exercise discretion over the timing of provisions for certain loan losses". Shen and Chih (2005) list three factors that demonstrate the importance of studying earnings management in the banking industry. First, banks at all times fear a potential problem of illiquidity that puts them under the risk of extensive bank runs. Therefore, with the intention of retaining depositors' confidence, banks resort to earnings management practices in order to avoid negative earnings.
Second, they cite Morgan (2002) who says that: "… uncertainty over the banks stems from their assets, loans and trading assets in particular, the risks of which are hard to observe or easy to change. Banks' high leverage compounds the uncertainty over their assets; their assets present bankers with ample opportunities' for risk or asset substitution, and their high leverage inclines them to do so." Therefore, bank managers have a high incentive to manage earnings to hide asset substitution behaviour.
Third, banks are highly regulated organizations in which a non-performing loan ratio, among other things (i.e. capital adequacy ratio, liquidity ratio, etc.) is firmly regulated. Therefore, earnings management could be adopted in order to avoid regulations' breach.
According to Kanagaretnam et al. (2010) a bank's LLP is the proper approach to study earnings management in the banking sector for two reasons. First, given the considerable discretion that is allowed by GAAP, bank managers may use this flexibility in using LLP for earnings management.
Second, LLPs are considered to be major accrual items in banks accounts that provide bank managers with sufficient leeway in manipulating earnings.
The literature shows a great deal of research on earnings management practices. However, only little is related to the banking industry (Taktak et al. 2010b). According to Peasnell et al. (2000, p. 318), financial institutions may be excluded due to the difference in the financial reporting system between financial firms and industrial firms. Also, financial firms have a "fundamentally different accrual process" relative to other industries.
According to Burgstahler andDichev (1997, cited in Shen andChih, 2005) conflicting incentives may exist within regulated firms in general to report lower earnings or decreases in earnings whenever economic benefits from reporting lower earnings to regulators take place. More particularly, in the financial institutions category which includes the banking industry, there may be a negative relationship between avoiding earnings decreases and the extent of regulatory oversight. Much research on financial institutions is carried out in the USA and Europe and has come to a conclusion that LLPs are widely used for, amongst other things, earnings management (Anandarajan et al. 2003(Anandarajan et al. , 2007Leventis et al., 2011).
Researchers have adopted different techniques to test for bank managers' use of LLPs to manage earnings. Some used the specific accrual technique; others have applied the earnings distribution techniques. The next section provides a summary of some studies that examined earnings management or specifically income smoothing as a form of earnings management within the banking industry. It is notable that none of these studies followed the first quantitative technique (total accruals). Bhat (1996) examined the income smoothing hypothesis for 148 banks that reported their earnings over the period 1981-1991. To examine whether banks smooth their income the researcher regressed logarithms of earnings after taxes and LLP against the year. His results suggest that banks use LLP to manipulate reported earnings.
Another study that followed the specific accrual technique was conducted by Kanagaretnam et al. (2003) who studied the underlying motives of bank managers for income smoothing through LLPs.
They provided evidence that bank managers use LLPs for income smoothing as they save earnings through LLPs in good times to borrow them in bad times. Their results suggest that job security and the cost of borrowing motivate bank managers to engage in income smoothing practices; managers faced with job security fears typically use LLPs as an income smoothing device. They claim their findings to be of great interest to regulatory bodies who are interested in banks' financial reporting quality. Their results show that Australian banks in general use LLPs to manage earnings, and, listed banks are more likely to do so relative to unlisted ones. Their results also suggest that regulators consider the fact that managed reported earnings do not precisely represent the real economic performance when assessing the "overall financial risk". Agarwal et al. (2007) investigate earnings management practices within Japanese banks and found out that Japanese banks do significantly use LLPs to manage earnings in the period of 1985-1996.
Other evidence from Japan is provided by Kwak et al. (2009)  The specific accrual technique has been also applied to Islamic banks. On a sample of 66 Islamic banks from various Muslim countries, Taktak et al. (2010b) found that Islamic banks extensively smooth their incomes. However, their study provided no evidence on the use of LLPs to smooth income by Islamic banks. They examined whether Islamic banks do smooth their income using LLPs. However, they studied only one form of income smoothing that is, artificial income smoothing through LLPs resulting from Islamic financial products i.e murabaha, musharaka and mudaraba.
Other studies, for example, Shen and Chih (2005) adopted the third quantitative technique that is, earning distribution discontinuities. In their study as to whether earnings management within the banking industry is practiced across 48 countries they found that earnings management certainly was practiced in their sample. They used three measures to test for earnings management based on those of Burgstahler and Dichev (1997) and Leuzet al. (2003). They tested whether banks manage their earnings so as to surpass thresholds, such zero earnings and zero earnings changes. Leventis et al. (2011) examined the impact of IFRS implementation on the use of LLPs to manage earnings within 91 EU listed commercial banks operating in 18 European countries. They divided their sample into two categories; early adopters and later adopters. Their results reveal that banks do manage their earnings through LLPs but the implementation of IFRS has meaningfully reduced earnings management behaviour.
The earnings management in Islamic banks has been also confirmed by using the earnings distribution approach. Hamdi and Zarai (2012)

Research methodology
This papers aims to investigate the perceptions of different stakeholders about the earnings management motivations and techniques in the LCBs. The empirical work of the current paper draws on findings concluded by Barghathi et al. (2017), as a result the questionnaire was designed in order to examine the motivations and techniques of earnings management in LCBs. It locates itself in the interpretive paradigm. In the interpretive paradigm, the researcher elicits the individuals' perceptions toward an issue as is done in this study. The results of this study will be discussed from the accountability perspective; in other words the conceptual framework of this study is accountability. In addition to the above aims; this study is focusing on the accountability relationships of banks managers with; shareholders and other stakeholders. By other words, this study tries to examine whether evident accountability relationships are taking place in the context of financial reporting of the Libyan Commercial Banks. Earnings management may have an adverse consequence on accountability relationships. Accountability, or being accountable, relies on managers providing useful, unbiased, and reliable information to the firm's stakeholders. Aers et al. (2013) stated that earnings management could reveal an accountability breach, they indicated: "Given users' ex ante uncertainty with regard to management's earnings management motives, indication of earnings management may be perceived as a significant accountability predicament, and bring management to offer more explanations on performance-related matters in " (p. 95).
A questionnaire should be pre-tested to ensure that all questions and instructions are clear to participants, it is also good practice to test the time it takes a participant to complete it. The first pilot study was undertaken in December 2012 in the School of Business at the University of Dundee on three PhD students who were undertaking their doctoral studies in different accounting topics. Their comments and ideas were important and resulted in some helpful suggestions. The second pilot study was undertaken in Libya with three auditors, one of whom holds an MSc in accounting from a US university and two academics, one of whom holds a PhD from an Egyptian university, to ensure an accurate translation and understandability of the questionnaire. The feedback was useful in terms of the Arabic translation and questions were modified based on the suggestions of two of the auditors.
Thus self-administration was selected, the distribution process took place in the period early January 2013 till February 2013 during which time 193 copies were given to various stakeholders of the Libyan Commercial Banks. The researcher had the benefit of a helpful network that had been developed over the previous years from the audit profession; a number of potential respondents were personally known to the researcher, or are currently, or were, clients for either audit or non-audit services of the office where the researcher works 1 . In addition, some of respondents offered to help to distribute more copies of the questionnaire to their own colleagues who worked in the same place and also to contacts in other banks. The procedure with other stakeholders was largely the same. Regulators from the Central Bank of Libya, for example, were accessed through a family member who works for the bank in a non-accounting position, as well as through the manager of the Benghazi branch of the central bank of Libya who is known to the researcher and who was of great assistance when asked to contact more employees to encourage their participation in the survey. Personal contacts were also useful in getting co-operation from the tax authority, the Libyan Stock Market, external auditors, academics and other stakeholders.
In total 193 questionnaires were distributed and collected personally. Questionnaires were attached to a covering letter stating the purpose of the study and encouraging the participants to take part in it; it also identified the researcher and assured the anonymity and confidentiality of any information which respondents would provide.
One of the most common ways to examine the scale of reliability of a questionnaire is Cronbach's. This test, according to Saunders et al. (2012), is used to measure the response consistency among the questions; its values range between 0 and 1.0 where a scale of 0.7 or above is considered as acceptable. The Cronbach's alpha for this study has been calculated by SPSS and the generated score was 0.922 which is, according to the above recommendations, acceptable.
As shown in Table 1 the response rate of this questionnaire was 53%. It varied, amongst the groups of respondents, from 48% for Preparers to 64% for Regulators.

Stakeholders' Perceptions about Earnings Management Motivations This section investigates respondents' opinions about the motivations that induce bank managers
to indulge in earnings management practices. Several questions have been listed in this section; these were mainly based on the literature. The questions were split into two groups (state and private) in order to examine whether ownership has any influence over perceived earnings management motivations in the LCBs; they also examined the influence of the Libyan Stock Market's effect on earnings management practices by asking if it was thought to affect the bank's share price. The results are shown in Table 3 (Panel A) and reveal a relatively strong level of agreement overall among stakeholder groups for all the questions. This suggests that the listing requirements were perceived to influence bank managers to engage in earnings management practices no matter the type of ownership as state and private banks were seen to be similarly motivated on this issue. More specifically, the Auditors group recorded a slight difference for question (Q 1.6) which investigated whether there was any institutional effect on bank managers that encouraged involvement in earnings management practices. They perceived that State owned banks, on average, were not motivated by other banks behaviour (mean score 2.92), and that private banks were, on balance, neutral on this motive; their mean score was 3.00. This result should be considered in light of the fact that most of the Auditors who responded (14 out of 27) have no professional qualification (see Table 2) which therefore, may explain this unexpected response. The questionnaire responses by Auditors only (Appendix xx) reveals that Auditors agreed and disagreed with this statement almost equally; as 7 agreed whereas another 7 disagreed and only one Auditor strongly disagreed with this statement. It seems that this Auditor, in addition, to a relatively large amount (11) who gave a neutral responses to this statement, is responsible for the below mid-point mean score (2.92) that was generated by the Auditors group.
As can be noted in Table 3 (Panel A), the KW test reveals no significant differences. Note: This table shows the mean and standard deviation (SD) for all respondents regarding questions about earnings management motivation in LCBs. It also provides the mean for each group and the p-value for the Kruskal-Wallis (K-W) test. Groups are defined as; preparers (PR), auditors (AD), regulators (RG), and users (US) for each question. A * indicates significance at the 5% level.
A 5-point Likert scale was used in these questions. It ranged from 1= "Strongly disagree" to 5= "Strongly agree".
The results reported above in Six MW tests were performed to identify any significant differences between any two pairs. The results of these tests are shown in Table 3 (Panel B). groups regarding questions about earnings management motivation in LCBs. Groups are defined as; preparers (PR), auditors (AD), regulators (RG), and users (US) for each question. A * indicates significance at the 5% level.
A 5-point Likert scale was used in these questions. It ranged from 1= "Strongly disagree" to 5= "Strongly agree".
The MW results reveal four significant differences across the different stakeholder groups. First of all, the Preparer and Auditor groups seem to have different views about whether bank managers practice earnings management in order to affect the share prices of private banks. Both groups agreed, on balance, that private LCBs are motivated to practice earnings management in order to influence their share price (Capital Market Motivations). However, a notable difference can be observed by looking at the questionnaire responses for each individual group. It is clear that Preparers stand on the side of agreement given that only 2 Preparers (out of 21 who responded to this statement) selected the disagreement option. On the other hand it is less obvious where Auditors are standing. Although 11 Auditors (out 25 who responded to this statement) expressed agreement with the statement, 2 of which strongly agreed, six (out of 25) disagreed with the view that private LCBs engage in earnings management (see Appendix xx). The reason why the Auditors' attitudes may not be that clear could be that 8 Auditors selected the neutral response.
The response of the Auditor group, in this study, should also be treated with caution in light of the fact that only 13 (out of 27, the total number of Auditors group) are professionally qualified. Therefore, some of Auditors (the remaining 14) may lack the expected experience to answer the questions. For example, they may not have dealt with this statement, and others, with appropriate 'professional scepticism' and therefore, their answers may not be consistent with the rest of the Auditor sample 2 .
The second significant difference relates to the Preparer and User groups regarding whether State owned banks manage earnings in order to avoid reporting losses. Again this difference should be looked at in light of how the questionnaires were answered by each group individually. Preparers showed, on average, a level of agreement with the view that state banks may engage in earnings management in order to avoid reporting losses with no a single Preparer expressing a strong level of disagreement with this statement. In addition, a relatively large number of Preparers did not respond to this statement (7 out of 27) and 6 responded neutrally. This leaves us with only 14 Preparers who showed, on balance, a clear overall attitude; 5 Preparers disagreed, another 5 agreed, and 4 strongly agreed with this statement. As for the Users group, they showed a clear level of agreement with this statement with only one response on the strongly disagree side. It should be noted that all Users responded to this statement; 17 agreed, 6 strongly agreed, and very few (4) selected a neutral response.
A third significant difference revealed by the MW tests was between the Regulator and User groups as to whether state banks engaged in earnings management in order to avoid any regulator intervention. The Regulators group answered this statement with a neutral mean score while Users were, on average, in agreement with a mean score of 3.68. This significant difference could be, partially, due to the ambiguous attitude of Regulators to this issue; as 6 agreed while only 5 disagreed with the statement, one of which showed strong disagreement. On the other hand, Users' attitudes regarding this statement were clearer in their agreement; the total number of Users who agreed to this statement is 20 including 4 who strongly agreed that state LCBs may be engaged in earnings management to avoid regulatory interventions.
Finally, a significant difference exists between the Regulator and User groups in terms of whether state banks are motivated to manage earnings in order to avoid report losses. The general view of Users in respect of this statement was clearer than the opinion of Regulators. The total number of Users who expressed a level of agreement with this statement is 23 of which 6 strongly agreed with the notion expressed. On the other hand, only 9 Regulators agreed with the statement (one of which strongly agreed) while a relatively significant number (4) of Regulators disagreed to this statement 2 of which strongly disagreed. There is a wide spread level of agreement amongst stakeholder groups that most of the listed motivations are perceived as potentially motivating LCBs to engage in earnings management practices. The Auditor who gave this statement a mean score of 2.92 (for Q 1.6.1 Table 3, Panel A) apparently did not believe, on average, that a public commercial bank would be inspired by other banks' behaviours to engage in earnings management. Also the Auditors group appeared neutral in relation to a similar question relating to private banks (Q 1.6.2 Table 3, Panel A) as their mean score was 3.00.
This general level of agreement could reflect stakeholder groups' beliefs that such motivations, when they exist, represent threats to the accountability relationship of LCBs regardless of whether they are public or private, or listed or unlisted. An LCB manager may not be seen to be properly accountable when such motivations exist.

Perceptions about Earnings Management Tools
This section identifies to what extent stakeholders agree regarding the techniques which might be used by bank managers to alter reported income. The respondents were asked to indicate their agreement or disagreement regarding the effectiveness of some potential earnings management techniques. The Kruskal-Wallis test was conducted, at first, to identify any significant differences between stakeholders groups. The test results are presented in Table 4 (Panel A). With few exceptions, most stakeholders groups' perceptions showed similar means, and most of them indicated their agreement with the potential usage of the earnings management methods listed in suggesting it as potentially the most likely method to be used by bank mangers to change their firms' reported income. It has to be mentioned that the term 'reserves' does include, inter alia, the Loan Loss Provision (LLP) which was mentioned a lot during the interviews. However, in this questionnaire the LLP is separately examined as will be discussed later.
The lowest mean score was given to foreign currency (Q 2.8) and this was because of different views held by the Auditors group regarding this question; apparently auditors, on balance, do not agree that foreign currency may be used for earnings management by LCBs as their mean score was only 2.88. Another two areas of disagreement were in Q 2.5 and Q 2.9 which both had the same overall mean score of 3.33. Concealing losses (Q 2.5) which means that bank managers may not register transactions in a timely fashion could have an adverse impact on the reliability of the reported income e.g. failing to book a huge amount of expenses in the current period. For this question, most stakeholders agreed in aggregate that such a tool could be used for earnings management; the exception was the Preparers group that was non-committal since it only gave this question a mean score of 3.00. The third area of disagreement was about the manipulation of the loan loss provision (LLP) (Q 2.9). In this question, the mean score of the Preparers group, on balance, suggests doubt about whether the LLP could be used for earnings management purposes; but all the other stakeholders groups agreed with the view that it could as indicated by the mean scores. However, an overview on the questionnaire responses by Preparers (Appendix xx) may be informative; the total number of Preparers who answered this question is 25 (out of 27) and the total number of those who agreed (11) is greater than those that did not agree (9) while 5 gave neutral answers. The reason why the overall mean was below 3.00 is that five of the nine Preparers who disagreed strongly disagreed that LLP could be used as an earnings management tool. As the previous Chapter mentioned, the LLP is one of the provisions a company usually maintains. It, on balance, highlighted by all stakeholder groups (according to the overall mean) to be one of the techniques that LCBs' manager may use since it received the highest mean score of 3.71. Note: This table shows the mean and standard deviation (SD) for all respondents regarding questions about earnings management tools. It also provides the mean for each group and the p-value for the Kruskal-Wallis (K-W) test. Groups are defined as; preparers (PR), auditors (AD), regulators (RG), and users (US) for each question. Bold figure indicates significance at the 5% level.
A 5-point Likert scale was used in these questions. It ranged from 1= "Strongly disagree" to 5= "Strongly agree".
LLP is suggested by the accounting literature to be amongst the most likely tools for earnings management by a bank. Moreover, it may be the most widely used technique since it represents a large accrual figure that bank managers can use for earnings management (Sun and Rath, 2010).
The results reported in Table 3 are consistent with the literature in respect of the usage of LLP as an earnings management tool.
The second highest mean score (3.68) was given to Investments (Q 2.7.4). The term "Investments" refers to either long term or short term amounts invested in shares. The valuation of these shares may increase or decrease at the year-end which requires an accounting treatment that would affect the reported income on a Fair Value basis. Based on this result, and given that it received the second highest mean score, one might argue that LCBs depend, to a large extent, on 'investment' to manage their earnings which is inconsistent with both the literature and prior interview findings.
Although the KW results did not reveal any significant differences in Table 4 (Panel A), to be consistent with the analysis order adopted earlier, Mann-Whitney tests were performed for more details and discussion. The results of the six MW tests are displayed in Table 4 (Panel B).  (PR), auditors (AD), regulators (RG), and users (US) for each question. Bold figure indicates significance at the 5% level. A 5-point Likert scale was used in these questions. It ranged from 1= "Strongly disagree" to 5= "Strongly agree".
The test results shown above in Panel B reveal three significant differences among the stakeholders.
The first significant difference is between the Preparer and Regulator groups as regard to the use of LLP as a means of managing LCBs' earnings. The Preparers' mean score was only 2.92 (the least) which suggests that these stakeholders do not regard this method as a potential way for managing earnings in the LCBs. On the other hand, the Regulators group showed a mean score of 3.67 (the highest) revealing their relatively strong agreement with the view that LLP may be used for earnings management by LCBs. In addition, the questionnaire responses by the Regulators group (Appendix xx) reveal more details that may clarify this difference in perceptions. The responses of Regulators to this statement were, to a large extent, clearly in favour of the potential use of the LLP, as 15 Regulators expressed agreement with the statement whereas only 2 expressed their disagreement, one of which was strong disagreement. The responses of the Regulators should be viewed in the light of the fact that most of them have work experience at the Central Bank of Libya and three of them work for the tax authority. Both these experiences may indicate a good deal of knowledge regarding the environment in which LCBs work and the way LCBs' financial information is compiled and presented.
On the other hand, Preparers' responses were not that clear. As discussed above, 11 Preparers agreed that LLP could be used to manage earnings, one of which was strongly agree, while 9 Preparers disagreed with the statement including 5 who strongly disagreed about the potential use of LLP for earnings management by LCBs.
The second significant difference appeared between the Preparer and User groups regarding the use of loss concealment (Q 2.5) by LCBs to affect reported income. This difference may have occurred because of that the Users group has had the highest mean score of 3.71 while the Preparers' gave more neutral responses yielding an exact mean score of 3.00. This apparent different can also be explained by looking at the questionnaire responses of individual groups. For example, the questionnaire responses from Preparers (reported in Appendix xx) shows that 10 Preparers agreed that concealing loss is potentially used to manage LCBs' earnings. On the other side, 8 Preparers disagreed with this view; 2 of them expressed their strong disagreement. 5 Users disagreed and 19 agreed including 6 who strongly agreed.
The last significant difference is that between the Auditors and Users groups regarding the usage of exchange rates manage LCBs' reported income. This was the subject of aggregate disagreement by the Auditors group according to their mean score of 2.88 (the least) compared to Users' mean score of 3.54 (the highest). This potential conflict in views can be explained as follows: most Auditors are not professionally qualified and thereby may arguably be viewed as lacking the required scepticism and/or it may simply be that some of the respondent Auditors do not possess a high level of experience, given the fact that 15 (55.6%) of Auditors have less than 5-years of banking experience. On the other hand, the Users group consisted mainly of academics that may, arguably, be viewed as having less practical experience but look at such issues normatively.
However, all other stakeholder groups showed, on balance, agreement that 'foreign currency' may be used by LCBs' managers to alter reported earnings.

Summary and Discussion
The motivations which are thought to be behind the exercise of earning management in LCBs are the same as those in the literature. Management compensation and job security, regulatory and political, and capital market motivations are perceived, on balance, by all stakeholder groups to have induced LCBs' managers to intervene in the process when it comes to determining the reported income. This can be achieved, according to the questionnaire results by all listed techniques that respondents were asked about. LLP, which the literature suggests is the most influential technique of earnings management by banks, is perceived to be used by LCBs' managers as well as other techniques. The results also suggest that LCBs' managers tend to use both accounting and real earnings management.
The evidence provided by the results refer to a serious problem to the accountability. The existence of these motivations, although, they may be unavoidable, put pressure on the accountability process and expose it to a lack of trust and disrepute which therefore may have an adverse effect on the relationship between LCBs' managers and their stakeholders. LCBs' manager should be free from such motivations in order to produce unbiased and fair accounting information.
This study is also concerned with how LCBs' managers could alter the reported earnings. As reported in the literature, LLP is quite commonly used by bank managers as a tool for earnings management. And apparently this is the case in Libya. The results clearly suggest that stakeholders agree, on balance, that LLP is being used by bank managers to alter their earnings, however, other techniques have been agreed, on balance, to be used as well.
Whatever the technique to manage earnings is to be, the main issue is the existence of earnings management. This raise concerns about the quality of financial reporting. The questionnaire results provide evidence on the existence of earnings management in LCBs' financial reporting which may refer to low quality and breach of accountability.
The financial reporting that is being provided to LCBs' stakeholders, based on the results of this study, may not be fair i.e. the objectivity of the accounting information is not properly applied.
Another major finding this study may refer to is that the questionnaire results provide evidence of LCBs' managers actually committing fraud when they thought of themselves as managing earnings. Concealing losses, for example, is perceived, on average, to be used by bank managers to manage earnings. This evidence may be seen as an explicit evidence that accounting information