There have been a large number of studies on perceptions of auditor's independence. Some examples can be Dykxhoorn & Sinning (1981) in German, Gul (1989) in New Zealand, Gul & Tsui (1992) and Lau & Ng; (1994) in Hong Kong, and Alleyne et al. (2006) in Barbados to name a few. There are only some published studies focussing mainly on the factors affecting auditor's independence (i.e Gul & Teoh, 1984; Teoh & Lim 1996; Abu Bakar et al. 2005, 2009).The study Gul and Teoh(1984) analyses the main effects of combined audit and management consulting services provided by public accounting firms and the population sample taken comprised of bankers, public accountants, mangers and shareholders. The result obtained was that the expansion by audit firms into non audit services reduced their confidence in the auditor's independence.
The study Teoh & Lim (1996) investigate the effects of five selected factors of AI of Malaysian public and nonpublic accountants. They make use of a repeated measures experimental design. The results conclude that a large audit fee received from a single client is the most essential factor leading to the risk of losing AI, followed by the provision of management consultancy services. The non-rotation of audit firms is not considered to be a dominant factor but the formation of audit committees is found to have a strong positive impact on improving auditor independence, while the positive impact of disclosure of non-audit fees is considerably less.
Some studies can be those of Abu Bakar(2005) who analyses the factors influencing auditor's independence from the perceptions of Malaysian loan officers. The study examines the perceptions of commercial loan officers who were relatively sophisticated financial statement users who would understand the importance of audit report and the issues related to auditor independence. A total of 86 officers responded to the self administered questionnaire. The results indicate that audit firms operating in a higher level of competitive environment, larger size of audit fees, audit firm serving a client over longer duration, audit firm providing managerial advisory services and non-existence of an audit committee are perceived as having the risk of losing auditor's independence. The most important factor affecting AI is given by Audit firm size, followed by tenure, competition, audit committee, MAS and size of audit fee. Another study by Abu Bakar (2009), attempts to explore the main determinants of auditor independence as perceived by Malaysian accountants. A self administered mail survey was used and a total of 72 completed questionnaires were received producing usable replies of 14.4%. From the survey, it is evidenced that, larger size of audit fees is the most important factor that is perceived as having risk of losing auditor independence followed by other factors [1] . Additionally, the study also provides a basis for the profession to establish policies relating to auditor independence.
Review of literature in terms of the different factors
In reality there are many factors which impair auditor independence and some studies concentrated on only one factor. For example, Salehi(2009) examined non audit services and audit independence. The result of this study strongly agrees that providing NAS to external auditors to the same client impair auditor independence. Several prior studies concluded that NAS has negative effects on auditor practices and auditor independence. A Survey carried out by Wines (1994) suggests that auditors receiving NAS fees are less likely to qualify their opinion than auditors that don't receive such fees, based on his empirical analysis of audit report issued between 1980 and 1989 by 76 companies publicity listed on Australian stock exchange. He found that auditors of companies with clean opinions received higher proportion of non audit fees than did auditors of companies with at least one qualification.
According to Beeler and Hunton (2002) contingent economic rents such as potential non-audit revenue, increase unintentional bias in the judgments of auditors. They found experimentally that audit partner participants searched more supportively, weighted confirmatory evidence more heavily, and made more elaborate arguments in the presence of low balling and potential non-audit revenue than provision of audit, and NAS claimed that auditors would not perform their audit service objectively and that joint provision would impair perceived independence (Glezen and Miller, 1985 [2] ;). Mitchel et al. (1993) believed that the joint provision of audit and NAS to audit clients would cause unfair competition due to the use of audit services to the same client and thus would impair AI.
Several prior studies also suggest that NAS has positive effects on auditor practices and auditor independence. Gul (1989) who studied the perceptions of bankers in New Zealand found that the effect of provision of NAS was significantly and positively associated with auditor independence. Hussey (1999) reported that the majority of the UK finance directors that participated in his study suggested that joint provision of audit and NAS to audit clients should continue to be allowed. In Malaysia Gul and Yap (1984) reported that NAS provision increased their confidence in auditor independence. Arruanda (1999, p. 165) pointed out that joint provision of audit and NAS would reduce overall cost, raises the technical quality of auditing, enhance competition. This would ultimately increase auditor independence. Kinney et al (2004) denoted knowledge of a client's information system and tax accounting could spill over to the audit, improve the information available to the auditor and thus improve audit quality which in turn would increase the probability that problems are discovered.
Audit committee
The author Sori (2009) made the study of audit Committee and Auditor Independence through the Bankers' Perception. The questionnaire and the interview survey reveal that the majority of the respondents agreed that auditor independence would be safeguarded by the presence of an active and independent audit committee. The audit committee is responsible for approving and reviewing audit fees as the majority of audit committee members are independent and non-executive directors. Teoh & Lim (1996) in their study find that the formation of audit committees has a strong positive impact on enhancing auditor independence. Similarly, Patten & Nuckols (1970), Knapp (1985) and Lau & Ng (1994) find that the existence of an audit committee increases the likelihood of bankers' approving a loan, which is a reflection of an increased confidence in the auditor. On the contrary, Gul (1989) suggests that audit committees did not significantly affect the perceptions of auditor independence.
Competition
Many empirical studies have proven that the high level of competition in the audit firm has resulted in less auditor independence (e.g. Shockley, 1981 [3] ;). However, Gul (1989) [4] who obtains the opposite, in explaining this, he argued that the existence of competition caused auditors to be more independent and create a favourable image in order to maintain their clientele. In a UK study (Beattie et al 1999) competition was the factor influencing auditor AI. The sample comprised of audit partners and the author argued that the factors affecting the perceptions of AI are likely to change over time owing to changes in the local economic, political, cultural and regulatory environment.
Size of audit firm and auditor tenure
Almost all empirical studies that attempted to find relationship between larger audit firm size and AI concluded that there is a positive relation between them [5] (De Angelo 1981). The author DR Zulkarnain (2006) analyses the size of audit firm and perceived auditor independence in his study. Questionnaires and interview survey were used to seek the perceptions of senior managers of audit firms, banks and public listed companies. The result concluded that the Big Four firms were perceived to be superior and more efficient compared to the non-big four firms in all aspects relating to independence from their clients. The respondents indicated that big four auditors are better able to resist management pressure in situations of conflicts and are more effective at detecting activities that will affect clients' company continuity. The non-big four firms are more risk averse with regard to litigation arising from fraud and irregularities compare to non-big four firms.
Most writers [6] , who discuss the relationship between tenure and AI, support that audit firms serving a given client over a longer duration has the risk of losing an auditor's independence. However, in studies conducted by Shockley(1981) and Teoh & Lim(1996) tenure was not found to have a significant impact on perceptions of AI.
Audit and non audit fees
Most empirical studies conducted on size of audit fees do not look at that factor itself; instead the studies are inter-related with other factors. For example, Shockley (1982) in his study suggests that the negative effects of MAS, the size of the audit firm and competition on a third party's PAI actually arise because of the linkage of these variables to audit fees.
However, there is a study that proves otherwise. For example, Gul (1991) who analyses banker's perceptions of AI proves that each independence-related variable such as the audit firm size, affects bankers' PAI in its own right. He also found size of audit fees to be essential factor of bankers' PAI. Another study related to the size of audit fees was by Pany & Reckers (1983).They noted that the large size of the client's audit fee (measured as a percentage of office revenues to the audit firm),though do not show any significant impacts on PAI, have influenced respondents to feel less confidence in the auditor's independence.
Additionally, where non-audit fees are concerned, several prior papers have studied the interests and importance of non-audit fees in terms of auditor independence. Unfortunately, the research stream that evaluates the association between non-audit services and auditor independence, by examining the effect of non-audit fees on the auditor's propensity to issue a going concern modified opinion (hereafter GC), has produced rather mixed results. For example, while in an international context the results are somewhat mixed and inconclusive (Lennox, 1999; Sharma, 2001; Sharma and Sidhu, 2001; Basioudis et al., 2008), contemporary researchers in the U.S. suggest that there is no relationship (DeFond et al., 2002; Geiger and Rama, 2003)1. Thus, despite the concerns of the regulators and the financial press, there is no clear evidence that higher non-audit fees negatively affect auditor independence.
Research on audit fees has also documented that client size is an important determinant of audit fees (Simunic, 1980; Francis, 1984), while recent research indicates that the relative magnitude of non-audit fees is also higher for larger clients (Abbott et al., 2002). Further, the provision of non-audit services by the incumbent auditor leads to better understanding of the client and knowledge spillovers (Simunic, 1984; Francis, 1984), and thus to a better informed audit reporting decisions. Together, these results suggest that audit opinions may be influenced by the magnitude of non-audit (and audit) fees received from clients.
Additionally, high non-audit fees, especially as a percentage of total auditor fees, may have a negative effect on stock market participants' perceptions of auditor independence. Prior research on the effect of non-audit fees on auditor independence is also inconclusive (for example, Wines (1994); Sharma (2001); DeFond et al. (2002); Basioudis et al. (2008), using GCs as proxy for auditor independence; and Frankel et al. (2002); Francis and Ke (2002); Ashbaugh et al. (2003); Chung and Kallapur (2003),using discretionary accruals as proxy for auditor independence).
In most empirical studies audit independence is proxied by the relative magnitude of the audit fee as against the NAF received from a particular audit client. Hoitash (2007) hypothesise that the fees paid to auditors can affect audit quality in two principal ways. First, large fees paid to auditors may increase the effort exerted by auditors and thereby increase audit quality.
