Independence in Financial Audit Engagements . The Case of Romanian Banks Univ

For the external audit to ensure the credibility of the financial statements, it is a mandatory and vital requirement to maintain the independence in audit engagements. In practice, when performing the audit engagements there are various threats to independence, such as the non-audit services, the financial dependence of the auditor on its client, and the familiarity between client and auditor. The risk they pose to independence may be diminished through the rotation policy of the partner and audit firm. The rotation (voluntary or compulsory of the partner or audit firm) generates, according to the professional literature, both advantages and disadvantages, the legal practices and regulations being different from one country to another. Although the Romanian legislation does not require the rotation within the audit engagements, we believe that adopting these practices is an indication of maintaining independence and conducting a quality audit. Starting from this aspect, our research aims at the rotation practices of the banking institutions operating on the Romanian market. The results show that most banks adopt their own rotation policy of the partner and audit firm. This research may be useful to interest-holders within the banking system, as well as to the regulatory bodies in Romania.


Introduction
In order for the financial audit engagements to be efficient and give reasonable assurance to the interestholders in respect of the credibility of the financial statements, it is compulsory to maintain the auditors" independence. The importance of the auditors" independence in performing the audit engagements is recognized in the professional literature, which brings up though the practices that may affect independence. To attenuate the risks posed to independence, the international regulations and practices introduce the rotation within audit engagements. The professional literature emphasizes the rotation advantages and disadvantages, both audit partner and the auditor, and the voluntary or compulsory rotation. Although there is no consensus in terms of the rotation practices that should be adopted in performing the audit engagements in order to ensure the audit independence and quality, starting from the professional literature, we consider that an optimum practice should aim the following aspects: the rotation of the audit firm should be performed after maximum 10 years and the one of the audit partner after maximum 5 years; the return of an audit partner to a client should occur only after a period of pause of minimum 2 years; the rotation of the firm and audit partner should be legally regulated; there should be imposed rules to forbid the client to change the auditor as a result of the audit opinion expressed by the auditor. Our research aims to analyze the independence while conducting the audit engagements within the banking institutions from Romania, by reviewing the adopted rotation policies. Although the Romanian regulations do not include specific requirements in respect of rotation, we consider that the presence of the rotation practices both of the partner and the audit firm is an indication in respect of maintaining the independence while performing the audit engagements, independence contributing to conducting a quality audit and giving more assurance to the interest-holders. The banking system was chosen to perform the research due to the fact that the banks have importance on the Romanian market and there is a wide category of interest-holders counting on the financial statements published by them. The current research is divided into 2 sections. The first part consists in the analysis of the audit firms" rotation in correlation with the issued audit opinion, and the second section takes into account the rotation of the audit partner within the engagements performed by the same firm of audit, following also the return policy of an audit partner after a certain period of time and the influence of the issued audit opinion. The results prove that, although the Romanian regulations do not display rules regarding the necessity of the firm or audit partner rotation within the audit engagements, more than 50% of the banks activating into the Romanian banking system implement their own rotation policy.
We believe that the results of the research will be useful to the interest-holders within the banking system, which is based on the quality of audit and independence of auditors, as well as to the regulating bodies from Romania which can take into consideration establishing certain rules to envisage the rotation of the partners and audit firms in terms of ensuring the independence while performing the audit engagements.
The rest of the paper is structured in the following manner: knowledge stage, research methodology, research results and conclusions.

Knowledge Stage
The economic development, free movement of capital, as well as the diversification of the categories of interestholders within a company, are only part of the elements emphasizing the importance of the external audit. The interest-holders, regardless of the category they belong to (management, shareholders, investors, banks, employees etc.), need a reasonable assurance that the financial statements presented by the company are in compliance with the legal regulations in force, they comply with the generally accepted accounting principles and do not represent the risk of material misstatements. The importance of such assurance increases if we take into account the big financial scandals which have shaken the confidence of the market and interest-holders and have contributed to the reform of audit practices and policies. In this context, the role of the financial audit or external audit becomes extremely important. External audit is therefore considered, according to the professional literature, a control mechanism, meant to protect shareholders and investors (Quick & Schmidt, 2018). The role of auditors is to provide assurances in respect of the correctness and credibility of the financial statements, reducing the information risk associated to them (Dart, 2011;Lu & Sivaramakrishnan, 2009).
In order for the external audit to be able to effectively fulfill its role, a very important element is the maintenance of independence. Independence involves reporting by the auditor of the errors encountered during the audit engagement (Quick & Schmidt, 2018;Dart, 2011). Without independence, it is considered that the audit will not detect or report possible errors or frauds that may occur in the financial statements. External audit can only perform its functions if the auditor guarantees the quality of the audit, which implies maintaining independence (Quick & Schmidt, 2018). Meuwissen & Quick (2019) state that protecting independence is the basis of the credibility of the audit profession. Beattie et al. (1999) consider the independence of the auditor as fundamental to maintaining confidence in the financial statements. In other words, if the independence of auditors is threatened, the credibility of financial statements will be questioned (Dart, 2011). Therefore, Chen et al. (2005) consider independence as the main element of the audit integrity.
The importance of independence in carrying out the audit engagement is recognized in the professional literature. According to the research conducted by Sarwoko & Agoes (2014) on the accounting firms registered on the Indonesian capital market, the auditor's independence and specialization in a particular industry have a significant impact on the audit procedures meant to detect fraud, and all three variables have a significant influence on the quality of the audit. Also, in Indonesia, Rahmina & Agoes (2014) demonstrate that auditor"s independence influences positively the audit quality. Therefore, the quality of audit is closely connected to independence. Without independence, the quality of audit is under question. In addition, Zhang et al. (2007) consider, based on the research conducted, that the independence of auditors also increases the probability of detecting the weaknesses related to internal control. The concept of independence takes into account the independence of both auditor and audit firm. Moreover, according to the literature, the auditor's independence must be both in fact and in appearance (Sarwoko & Agoes, 2014;Beattie et al., 1999;Ghosh et al., 2009). The independence in fact refers to the mental attitude of the auditor, who must be unbiased, whereas the independence in appearance translates into the public opinion or perception about the auditor's objectivity and the lack of a conflict of interests between the auditor and its audit client (Beattie et al., 1999;Ghosh et al., 2009). However, while the professional literature considers the independence of auditors as a central element in the performance of the audit engagements, Patel & Psaros (2000) consider that the implementation of a unique set at global level for the audit and professional conduct procedures is questioned due to the fact that, according to the research conducted by the authors, the independence of the external auditors is perceived differently by the students from the United Kingdom, Australia, India and Malaysia, depending on the country which they come from. Their research reveals that the perceptions on independence are similar for the countries that have economic, political and cultural connections. Therefore, we consider that the notion of independence in audit should be addressed according to the specific context of each country. However, regardless of the context and the possibility of adopting a unique set of audit procedures and rules of professional conduct, only one aspect remains invariable, namely the importance of maintaining independence in audit engagements.
Although in theory the concept of independence is easy to be described and understood and it is a central element that ensures or guarantees the success of an external audit engagement, maintaining independence in practice is a difficult task to accomplish due to many threats the auditor faces during its engagements.
Among the obstacles which may threaten the maintenance of the auditors" independence there are, according to the professional literature, the non-audit services provided by the audit firm simultaneously with the audit of financial statements, the familiarity that may arise between the auditors and the employer as a result of a long-term audit relationship and the economic dependence of the audit firm on the client (Meuwissen & Quick, 2019;Beattie et al., 1999;Ghosh et al., 2009). The first two situations strengthen the connection between the audit firm and the audited company, emphasizing the financial dependence and thus questioning the independence of auditors.
The auditors' financial dependence on their clients can be perceived as a problem affecting their independence. According to the study conducted by Booker (2018), creditors consider the financial statements of the companies with financially dependent auditors less credible in comparison with the companies whose auditors are not financially dependent on them. However, the credibility of the financial statements can be improved if companies voluntarily adopt a rotation policy of the audit firm.
In respect of the problems generated by the provision of non-audit services, which may affect the independence of auditors, they are related to: creating a financial dependence situation of the audit firm on the audited company, which can lead to ignoring certain errors due to the desire of the audit firm to keep its clients which bring financial benefits; the likely tendency of the auditors to ignore the errors which have arisen or been generated following the consulting services offered; as well as the familiarity connection that is created between the auditor and the audited firm (Meuwissen & Quick, 2019).
In the professional literature there are various studies analyzing the impact of non-audit services on the independence of auditors. The experiment carried out by Meuwissen & Quick (2019) in the particular case of the listed companies from Germany, takes into account the perception of the supervisory board members of the firms on the independence of auditors when providing for the same company both audit services, and nonaudit services (tax advice, counselling on the financial information system and human resources counselling). The results show that the provision of non-audit services generates a negative impact on the independence of auditors, especially the provision of counselling services in the field of human resources (although, unlike the first two, they are not prohibited by the German regulations). Also, the results of the study conducted by Dart (2011) indicate the fact that the economic dependence of the auditor on an audit client and the provision of non-audit services are considered by the investors in the United Kingdom to be greater threats to the independence of the auditors than the long-term relationship between client and auditor. Chen et al. (2005) also consider, following the research conducted in Taiwan, that the provision of non-audit services contributes to reducing independence, being found a negative correlation between the duration of non-audit services and the degree to which the client agrees with the auditors' opinion in respect of the financial reporting. In contrast, the fee for the non-audit services does not influence the auditor's decision when this one is employed for a longer period (more than 5 years).
Alternatively, following a laboratory study, Reckers & Robinson (2007) consider that the non-audit services do not compromise the independence in fact of auditors. Ghosh et al. (2009) do not find any evidence either to support the correlation between the independence of the auditors and the commission rate out of the total for nonaudit services. Moreover, it seems that companies are tempted to turn to other companies for non-audit services and not to the company that performs the audit. Albring et al. (2014) discuss the provision of tax advice services by auditors and their independence. They demonstrate through the performed study that the probability for the companies to use tax consultants outside the audit firms is positively associated with the efficiency of the audit committee and strong corporate governance rules, the decision being influenced by the manner in which it is perceived the auditors' independence and the quality of audit.
In order to protect the independence in case the auditor provides both audit and non-audit services to the same client, the European audit rules envisage the prohibition of providing certain types of non-audit services, the approval of providing non-audit services by the Audit Committee, as well as limiting the revenues which can be obtained from services other than the audit ones (Meuwissen & Quick, 2019). In order to improve the independence of auditors, The Securities and Exchange Commission prohibits the provision of certain non-audit services to the audited clients (Albring et al., 2014).
However, we consider that the rotation of auditors would allow the limitation of the risks that the provision of nonaudit services can generate by: limiting the possibility of creating familiarity connections with the audit client, limiting the provision of long-term non-audit services and limiting the possibility for the auditor to evaluate the results of the services provided. In terms of the impact of a long-term relationship between the auditor / audit firm and the client on the independence and quality of audit, the opinions are divided. According to the literature, a long-term commitment in carrying out the audit engagements presents both advantages and disadvantages. The disadvantages of maintaining a long-term relationship include: creating a feeling of familiarity between the auditor and the client that can reduce the professional skepticism of the auditor, who has a higher confidence in the audited firm; the tendency of the auditor to rely too much on the verifications previously made, which may lead to the reduction of the efforts made in the audit engagement; and so on (Quick & Schmidt, 2018;Ball et al., 2015).
In order to prevent the familiarity generated by the longterm relationships between the auditors and the audit client and the disadvantages mentioned above, there has appeared a solution in practice, namely the rotation. The rotation may refer to both the rotation of the audit partners and the rotation of the audit firm and it may be voluntary or compulsory. According to the professional literature, the advantages of limiting the relationship between the auditors and the client envisage: reducing the financial dependence of the auditor on the audited firm; increasing the independence of the auditors and improving the quality of audit; the advantage of the professional skepticism of another auditor and a new perspective on the audited elements (Quick & Schmidt, 2018;Ball et al., 2015). The rotation, however, diminishes the advantages that the long-term relationship between the auditor and the client can generate. According to the literature, a longterm relationship contributes to improving the quality of the audit due to the fact that it allows the auditor to detect errors more easily as a result of his / her deep knowledge and understanding of the audit client and causes the auditor to invest more in the knowledge and understanding of the audited firm, while limiting the cases in which the auditor is forced to give up independence in order to find new clients (Quick & Schmidt, 2018;Ball et al., 2015). The rotation or change of the auditors determines the increase of competition on the audit market, which can negatively affect the independence of the auditors who may be tempted to offer a favorable audit opinion in order to obtain an audit client (Choi et al., 2015). Moreover, the compulsory rotation of auditors can lead to increasing the implementation costs for the client and the auditor, diminishing the auditors' motivation to invest and to get a specialization in certain industries (Catanach & Walker, 1999). In this context, several questions arise. Is rotation in external audit engagements useful or not? Can rotation contribute to improving the quality of audit and the degree of independence, reducing the financial dependence of the auditor on a particular client, and reducing the negative effects that the provision of nonaudit services can generate? Is rotation in the audit engagements the magic answer to all threats that affect independence? And in the context in which rotation is an answer to the problems arising from independence, is the rotation of the company or auditor required? Should the rotation be voluntary or required by legal regulations?
In order to ensure the quality of audit, the European Parliament demands the rotation of the audit firm every 10 years and, by way of derogation, every 24 years when it comes to a joint audit, as well as a pause period of at least 4 years until the take-over of the same client  2015) on the companies listed on the stock exchange in Australia which implement IFRS follows the link between maintaining both the audit partner and the audit firm and the quality of audit. The results of the study indicate a negative correlation between the duration of the audit partner's maintenance (based on the relationship between the audit partner and the client's representative) and the quality of audit, but a positive correlation between maintaining the audit firm (based on the relationship audit firm -client firm) and the quality of audit. According to the authors, this aspect indicates the rotation benefits to the audit partner and not to the company that performs the audit service. Kaplan & Mauldin (2008) also analyze the perception of nonprofessional investors in terms of independence in the case of the rotation of the audit partners and the audit firm. The results do not indicate a different perception on independence, regardless of whether the rotation involves the change of the audit partner or the audit firm, but the confidence in independence is higher when the Audit Committee is powerful. Thus, the authors advocate for the rotation of the audit partner and not the audit firm. The results of the study conducted by Krishnan & Zhang (2019) also indicate that the rotation of the audit partner causes changes in how investors perceive the quality of audit. Therefore, changing the audit partner reduces the cost of equity, improves the earnings informativeness, and short-sellers consider the information related to earnings as bearing a higher quality after the rotation of the audit partner, the risk of audit being thus reduced as a result of rotation. However, the authors draw attention to the fact that, within complex companies, where the learning process is long lasting, the change of the audit partner adversely affects the quality of audit. Following the mandatory rotation of partner auditors in China every five years, the study conducted by Bandyopadhyay et al. (2014) indicates that, during the first three years after the rotation, the quality of audit is improved. A detailed analysis reveals the fact that the improvement is present in those provinces in China where legal regulations are poorly developed and the concentration of the audit market is low (high competition on the audit market). In the other cases, the rotation of the audit partner does not affect the quality of audit, where it was already high. Thus, the authors believe that the efficiency of the rotation of the audit partner depends widely on the structure of the audit market, too. Another aspect to consider when talking about the rotation of the audit partners refers to the required period after an auditor can return to a previously audited client as the main audit partner. Firth et al. (2012) analyze, in the particular case of China, when the period in which an audit partner can return to a client is of two years. The conclusions reached by the authors indicate the fact that the partners who return to a company tend to treat the client in a more favorable manner and consider that imposing a longer period for an auditor to return to a previously audited client is more beneficial for the quality of audit.
The rotation of the audit firms is also subject to analysis in the professional literature. Arel et al. (2006) bring to the fore a beneficial effect of the rotation of the audit firm. Thus, their experimental study reveals that the probability of changing the audit report to reflect the reality and the deviations from the generally accepted accounting principles is higher if the auditors expect rotation. The authors consider that the independence of the audit firm is thus increased under the conditions in which it is possible to talk about rotation. Daniels & Booker (2011) conduct a study on the opinion of the credit officers regarding the impact of the rotation of the audit firm on the auditor's independence and the quality of audit. The results show that when companies adopt a rotation policy of the audit firm, confidence in the independence of the auditors increases. However, the length of the audit mandate does not influence the opinion on independence. Quick & Schmidt (2018) analyze the impact which the rotation of the audit firm, the maintenance of the auditors and the joint audit have on the independence of the auditors and the quality of audit, analyzing the point of view of the bank managers and investors in Germany. The results show that the independence of the auditors is negatively influenced only by the joint audit, as well as the connection between the rotation of the audit firm and the joint audit. In contrast, the quality of audit is adversely affected by a 24-year rotation cycle, as compared to a 10-year rotation cycle and the interaction between a 24-year rotation cycle and the auditor's maintenance.
Imposing the changing of auditor by legal regulations can generate benefits. Thus, Wang & Tuttle (2009) consider, based on the conducted study, that the compulsory rotation of the audit firms causes them to adopt negotiation strategies less cooperative with the potential clients and impose their preferences. We can say that in this situation the influence of the client on the auditor and the risk of being affected independence are diminished. The study conducted by Corbella et al. (2015), in the particular case of the companies in Italy, where the rotation of the audit firm is mandatory, reveals the fact that the rotation of the company improves the quality of audit when this is performed by audit firms which do not belong to the Big 4 category and determines a decrease of the fees paid to the auditors within the Big 4 companies. Choi et al. (2015) analyze the effect of the auditor's change on the stock liquidity. They consider the particular case of South Korea, where the auditor is appointed by the Securities Commission for certain companies considered risky. Thus, in this case, the change of the external auditor is compulsory, not voluntary, which diminishes the risk of auditor manipulation by the company. Their study compares the impact that the voluntary or compulsory change of the auditor has on the market by referring to the liquidity of the shares. The research results indicate that the companies which have changed the auditor have a lower liquidity compared to those that have not done so, the effect being more powerful for the companies that move to an auditor that does not belong to the Big 4. The negative impact is diminished when the change is required. Moreover, differences in liquidity are insignificant in the comparative case of companies that do not change the auditor and those which have a change imposed by the regulatory bodies. In addition, it is ascertain that the negative impact of the voluntary change of the auditor is diminished if there is a high percentage of foreign ownership within the company. Casterella & Johnsto (2013) also address the problem of the efficiency of the imposed or compulsory rotation of the audit firm, analyzing the previously published studies. According to the authors, the results on the efficiency of the mandatory change depend on the data used in the analyzed researches, respectively a voluntary or compulsory rotation. The basic question is whether the effects of the voluntary change will be the same when the audit firm change shall be required. However, the rotation within the audit engagements can be used for purposes other than to ensure the independence and quality of audit. Heliodoro et al. (2016) emphasize an extremely important topic, namely the connection between the audit opinion and the auditor's change. This connection cannot be ignored, as it raises an alarm signal on the auditor's independence and objectivity. Thus, if the audit firm feels the financial pressure that comes from threatening the client to change the provider of the audit services, there may arise situations when the audit opinion is changed in favor of the client or in accordance with its requirements. This aspect affects the independence of auditors and questions the quality of audit and the credibility of the financial statements. The study carried out by Heliodoro et al. (2016) on the companies listed on the financial market in Portugal, shows a connection between the audit report and the change of auditors for the opinions regarding assets and others, but not for those related to capital and debts. The authors draw attention to the fact that companies may resort to the intentional change of the auditor to avoid presenting an unfavorable image of the company to the users of the financial statements. The study by Salleh & Jasmani (2014) on the Malaysian companies, where the rotation of the audit firm is voluntary, and the rotation of the auditor is mandatory, indicates that the audit opinion does not influence the rotation of the partner, but rather the rotation of the audit firm. This aspect draws attention on another problem that may occur in practice. How regulated should the auditor change be to allow the market to react to the auditor's change? Ferguson et al. (2018) discuss this aspect in their work. If the notification of the auditor's change is regulated and depends on the bodies which govern the stock exchange (as it is the case of Australia, where there is required the approval of the Australian Securities and Investments Commission for the public information on the auditor's change and it involves some constraints), there are advantages in terms of the uniformity of information dissemination and the approval only of justified changes. In contrast, if the regulatory bodies play a limited role, the information on the auditor's change is streamlined faster to users, thus increasing its usefulness, but at the same time leaving room for changes of auditors for opportunistic reasons.
According to the study carried out by Ferguson et al. (2018) on the market reaction to the notification of the auditor's change, in the particular case of the companies listed on the Australian stock exchange, which goes from an accentuated regulation to more flexible rules regarding the replacement of the auditor, a lack of evidence of the market reaction can be noticed to the companies" notification regarding the change of the auditor when this is regulated and a market reaction in response to the quality of audit when the change involves the audit firms belonging to the Big 4 category. Therefore, we note that, although the opinions on the importance of independence are consistent in the professional literature, the opinions on rotation within audit engagements show significant variations.

Research Methodology
The analysis of the professional literature reveals three important aspects. The first one refers to the importance of external audit for ensuring the credibility of financial statements. The second considers the importance of independence in carrying out the audit engagements, and the last aspect highlights the difficulty of establishing general rules regarding the rotation within the audit engagements. In Romania there are no regulations to envisage the rotation of the firm or audit partner at certain periods of time. Therefore, we consider (in our opinion), starting from the practices found in the professional literature and the benefits generated by them, that: the rotation of the audit partner after 5 years and the audit firm after 10 years can generate benefits for the audit independence and quality; in order to maintain the independence of the auditors and limit their financial dependence by the audited client, the rotation must be mandatory, by including it in the legal regulations; the auditor's voluntary rotation must be justified by the audit client in order to avoid the opportunistic changes or attempts to influence the auditor's opinion; an audit partner is not allowed to return to a previously audited client before a period of pause of at least 2 years. The present research aims at analyzing the independence within the Romanian banking system, by evaluating the voluntary rotation of the audit firm and the auditor during the annual audit engagements. Starting from the conclusions obtained from the analysis of the professional literature, we will follow, for all the banks operating on the Romanian market, the change of the auditor after 5 years at most and the audit firm after 10 years at most. We will also look at the period when an audit partner can return to an audit client (which we consider to be ideally not less than two years). We consider that the voluntary adoption of a rotation policy of the audit partner and/or the audit firm may indicate the concern of the company to maintain independence in the audit engagements. Two other issues that will be pursued in the current research take into account the country of origin of the banks and the audit opinion. The purpose of this analysis is to track whether the change of the partner or audit firm occurs after issuing an unfavorable opinion and to see whether the country of origin (where the rotation can be legally regulated) influences or not the adopted rotation policies.
In order to achieve the mentioned objectives, the list of banks operating in Romania was obtained from the website of the National Bank of Romania (https://www.cursbnr.ro/lista-banci) and updated and supplemented with the information available in the virtual environment. Thus, in the end, there was created a list with 29 banks currently operating within the banking system in Romania. Out of these, 5 are from Romania, 4 are from Greece, 3 are from Austria, France, Italy and the Netherlands and one from each of the following countries: Belgium, Germany, the United Kingdom, Israel, Poland, the United States, Turkey and Hungary.
For the 29 banks in Romania, there were analyzed, where available, the audit reports from 12 consecutive years, respectively for the period 2007-2018, following the three mentioned aspects regarding the rotation in the audit engagements (the audit firm, the audit partner, the return of an audit partner), as well as the audit opinion issued. Data collection, analysis and processing were performed manually. For a start, the official sites of the banks were searched and the audit reports were looked up in respect of the individual or consolidated financial statements. The information not available on the banks' official website was completed, where possible, with manually collected data available in the online environment.
Out of the 29 banks under analysis there were initially pursued for carrying out the research those for which could be confirmed the rotation policy of the company after maximum 10 years and then, the banks for which could be confirmed the rotation policy of the audit partner responsible for the audit engagement within the same company after maximum 5 years. Subsequently, due to the low number of observations, within the performed research there were also mentioned and analyzed the banking institutions bearing rotational indications, but for which the period the change (of the firm or the audit partner) takes place cannot be determined.
Only 4 banks out of the 29 banking institutions operating on the Romanian market (13.79%) display data on the audit firm permitting the verification of the rotation of the audit firm after a period of maximum 10 years. Another 13 out of the 29 banks (44.83%) have indications of the rotation of the audit firm (there is an obvious change of the audit firm), but it cannot be determined the time period after which the rotation takes place. For 5 banks (17.24%) there is not enough data to confirm or deny the existence of the rotation of the company that performs the audit mission, and for the remaining 7 banking institutions (24.14%) the data necessary to perform an analysis could not be collected.
In respect of the possibility of researching the rotation of the audit partner, out of the 29 banks, 8 (27.59%) change the main audit partner after a period of at most 5 consecutive years, 7 of them (24.14%) ) display the audit partner's rotation indicators, but the lack of data available for the previous and subsequent years makes it impossible to estimate the audit partner's rotation period, 5 banks (17.24%) maintain an audit partner functioning for a period longer than 5 years in a row, and for 2 of the analyzed banking institutions (6.90%) it is neither possible to confirm nor to deny the existence of a change of the audit partner. For the remaining 7 banking institutions (24.14%), the data required to perform an analysis could not be collected. It should be mentioned that for 6 out of the 29 banks the audit reports were not found on the official websites of the banks or in the online environment and for 1 bank only an audit report was found (about 7 banks or 24.14% out of the total of banks without published information on the audit report). Another important aspect to note is the fact that all the audit reports analyzed, for all banks and periods under review, were prepared by one of the Big 4 audit companies: KPMG, Ernst & Young, Deloitte or PricewaterhouseCoopers. Out of the total of the audited reports analyzed (184 audit reports), only 5 presented an opinion with reservations and only 1 audit report presented the impossibility of expressing an opinion.
In order to avoid identifying the banks subject to verification in the present research, they have been included in the study based on a random order, and the mentions regarding them will be made based on numbers assigned to each bank.

Results of the research
The first analysis in this study concerns the rotation of audit firms, in correlation with the issued audit opinion. Out of the 29 institutions operating in the banking system in Romania, only for 4 of them (13.79%) can the rotation of the audit firm be confirmed for a period shorter than 10 years. We mention that out of the 4 banks, only 2 have data regarding the audit process over a period longer than 11 consecutive years.
Bank no. 1 under analysis, presents information for 11 consecutive years. The institution is at first audited by EY then, for 8 consecutive years, the auditor is KPMG, after which, there is performed again a rotation of the audit firm towards Deloitte. It can be noticed that the period of keeping the auditor is shorter than the maximum limit of 10 years, the rotation of the audit firm taking place after a period of only 8 years. This aspect supports the idea of maintaining independence, especially in the context in which, in the situation of changing audit firms, the opinions were without reservations.
In the case of Bank no. 3, the available data extend over the entire period under analysis, respectively 12 years. The rotation of the audit firm is performed in the first instance from the company EY to Deloitte. The Deloitte auditor is maintained for 5 years after which a new rotation takes place, back to the audit firm EY. In this case the rotation takes place also after a period shorter than 10 years, increasing the confidence in the independence of the audit firm. Another important observation that emerges from our analysis is that even in this case, the change was not influenced by the audit opinion.
For Bank no. 11, data were available only for 9 consecutive years. However, an analysis of the rotation of the audit firm is possible. The bank replaces the audit firm KPMG with the audit firm Deloitte which it keeps for 6 years, after which it carries out a new rotation at KPMG. The rotation of the audit firm, performed for a period of less than the 10-year limit considered optimal for ensuring independence, is not influenced by the audit opinions issued by the audit firms.
Bank no. 16, although it has data available for only 7 consecutive years, has the most intense rotation of the audit firm. Thus, it makes the transition from EY to KPMG which becomes auditor within the institution for only 2 years, after which a new rotation is made to the company Deloitte, maintained in office for 2 years, and then it is followed by the PwC. In case the rotation of the audit firm can be noticed at a period of 2 years, the maintenance of independence can be confirmed. An observation that should be mentioned is that, after an audit report concluded with the impossibility of expressing an opinion, a rotation of the audit firm takes place, but, in the context in which the respective banking institution seems to follow a 2-year rotation policy of the audit firm, we do not consider this as an alarm signal on the independence of the audit or the opportunistic change of the audit firm.
Other 13 banks (44.83%) present indications regarding the rotation of the audit firm, but without being able to determine the period after which the rotation takes place due to the lack of the necessary information from the periods before and after the analyzed years. However, as a rotation of the audit firm is evident, we consider this aspect to be a concern for maintaining independence. We mention that out of the 13 banks, 2 have data for a period lasting more than 11 years.
Bank no. 2 provides information on the audit firm for a period of 12 years. During the first 10 years the audit was carried out by EY, and in the last 2 years by the PwC. Although a rotation of the audit firm is evident, rotation which is not influenced by the audit opinion (for all the reference years the opinion is without reservations), the period after which the rotation takes place cannot be identified. The EY auditor could have been maintained also during the previous period of the performed analysis (before 2007), which would mean that the rotation took place after a period longer than 10 years. For Bank no. 5 there was collected information regarding the audit engagement for only 8 consecutive years. During the analyzed years, the institution is audited for 2 consecutive years by KPMG (we do not have information about the audit firm from the previous period) and then by Deloitte for 6 years. Bank no. 6 displays information on the audit engagements for 10 consecutive years, and we could notice a rotation from EY (4 years) to KPMG (6 years). Bank no. 7, for which there are available data for 9 consecutive years, makes a rotation from the audit firm Deloitte (7 years) to the PwC (2 years). In the case of Bank no. 8, which has 4 consecutive years available for analysis, the rotation takes place from KPMG (1 year) to PwC (3 years). Bank no. 9 has data available for 6 consecutive years, the rotation being made from PwC to KPMG (4 years). For Bank no. 10, the data from the last 4 consecutive years show a shift from KPMG to EY (2 years). In the case of Bank no. 12, the data available for 4 consecutive years indicate a change of the auditor from Deloitte (3 years) to KPMG. Bank no. 13, with data available for 5 consecutive years, makes a rotation of the audit firm from PwC (4 years) to Deloitte (1 year). Bank no. 14 turns from KPMG (2 years) to PwC (5 years), and Bank no. 15 from KPMG (4 years) to PwC (2 years). For Bank no. 4, although it presents data for 12 years, there is only possible to confirm the rotation of the audit firm, without being able to verify whether it has taken place after 10 consecutive years. The banking institution was audited for the first 5 years by KPMG and then by EY for the next 7 years. The audit opinions expressed in the 12 years on the financial statements were without reservations. Bank no. 17 also carries out a rotation of the audit firm from EY (4 years) to KPMG (6 years).
None of these cases brings any evidence to support the fact that the rotation of the audit firm was influenced by the issued audit opinion. Moreover, in the case of Bank no. 9, the change of the auditor turns from an opinion without reservations to an audit opinion with reservations.
For the rest of the banks presenting data (5 banking institutions, respectively banks 18,19,20,21,22), the information found was not sufficient to confirm or deny the rotation of the audit firm after a period of maximum 10 years or even the existence of a rotation (e.g.: data available only for a period of 3 or 9 consecutive years in which the audit is performed by the same audit firm).
The results of the study are schematically presented in

Source: Authors" projection
The results of the research support the fact that within the banking system in Romania, there is present the rotation of the audit firms (for 17 analyzed banks for which there could be collected data from the 29 existing ones, about 58.62%) and it is not influenced by the issued audit opinion. These aspects present indications that support the independence of the audit firms which carry out audit engagements for the Romanian banking institutions. The second analysis carried out considers the rotation of the audit partner within the engagements conducted by the same audit firm, also being pursued the policy of an audit partner returning after a certain period and the influence of the issued audit opinion. Out of the 29 institutions present in the banking system in Romania, only 8 of them (27.59%) change the main audit partner after a period of at most 5 consecutive years.
Bank no. 1 is audited in the last 2 years by the same audit firm, for each of the two years having a different audit partner. The change of the partner was not made as a result of the issued audit opinion and, from the available information, there is no come back of the partners for a period shorter than 9 years, during which the audit was carried out by other audit firms. Bank no. 5 presents, in the 6 consecutive years of audit carried out by the same company, a change of the audit partner after 5 years. In the case of Bank no. 3, there could be noticed that, during the 5 years of consecutive audit performed by the same company, an audit partner is maintained in office for maximum 4 years. For Bank no. 9 during the last 4 years audited by the same audit firm, there are 2 audit partners maintained in office for 1 year, respectively 3 years. Therefore, the change takes place after a period of less than one year. However, the change comes after expressing an opinion with reservations, which may indicate a pressure exerted by the client on the audit firm. We still consider that this hypothesis is not valid in this case, the next report issued by another audit partner being issued also with reservations. Bank no. 11, in the 5 consecutive years analyzed, has an audit partner for 4 years, until the audit firm is changed. For Bank no. 13, during the last 4 consecutive years audited by the same audit firm, there are 3 different partners, out of which the last two (until the change of the audit firm) were kept for 2 years, respectively one year. For Bank no. 14, after the change of the audit firm, in the next 5 years 2 different audit partners can be identified, the first one being kept for 2 years. For Bank no. 15 also, in the last 4 consecutive years until the change of the audit firm, 2 auditors are identified, the latter having a mandate of 2 years. For all the banks analyzed, there is noticed the fact that neither the changes are influenced by the issued audit opinion nor there can be the case of the return of a partner after a period shorter than 2 years (where the available information allowed to analyze this aspect). Out of the 29 banks under analysis, the rotation of the audit partner is present in 7 of them (24.14%), but the lack of data available for the previous and following years makes it impossible to estimate the true rotation period of the audit partner.
For Bank no. 18, a rotation of the audit partner is noticed in the two consecutive years under analysis for which we have information. However, it is impossible to determine the period during which the previous audit partner was maintained, but we do not think that it can be the case of a return of a partner to the same client for a period shorter than 2 years. We can mention that the change is not influenced by the issued audit opinion. The same situation is common for the case of other banks. Thus, in the case of Bank no. 8, in only 6 consecutive years available for analysis, there is noticed a change of the audit partner, the last one being maintained for 5 years.
However, there is no information available for a later period to analyze whether the audit partner is kept also in the 6th consecutive year. For Bank no. 19 there is also a change of the auditor in the 8 years under analysis, but the lack of information for the previous and subsequent periods does not allow the verification of the hypothesis according to which the change is made after a maximum of 5 years. The same situation is true for Bank no. 20 (the analysis being performed for 4 years, in which there are 2 different partners), Bank no. 10 (2 consecutive years, with 2 different auditors), Bank no. 21 (3 consecutive years, with 2 different auditors), Bank no. 22 (an audit firm is maintained for 5 years, during which there are 2 different audit partners). For all the banks analyzed, there is noticed the fact that neither the changes are influenced by the issued audit opinion nor there can be the case of the return of a partner after a period of less than 2 years (where the available information allowed to analyze this aspect).
For 5 banks out of the analyzed ones (17.24%) there can be seen that an audit partner is maintained in function for a period longer than 5 consecutive years. For Bank no. 2, during the 9 years of consecutive audit carried out by an audit firm, there can be noticed that a partner is maintained in conducting the audit engagements for a period longer than 5 years, respectively for 7 consecutive years. When the partner change occurs, the audit opinion does not seem to be an influencing factor. We also mention that the new audit partner is different from those of the last 8 years, not even in this case being about the return of a partner. In the case of Bank no. 6, during the last 6 consecutive years analyzed, the audit partner was not changed. Bank no. 7 maintains, from the same audit firm, an audit partner for a period of 6 consecutive years. The same fact can be noticed within Bank no. 4, where the audit partner has a mandate of 7 consecutive years and in the case of Bank no. 17 where a partner is maintained for 6 consecutive years. For 2 banking institutions out of the 29 (6.90%) it is not possible either to confirm or to deny the existence of a change of the audit partner. Among these there are Bank no. 12 (for which there are data from the same audit firm for only 3 consecutive years in which the audit is performed by the same partner) and Bank no. 16 (2 audit firms are maintained for 2 years, during which time the audit partner is not changed).
The results of the study are schematically presented in Figure no. 2.

Figure no. 2. The rotation of the audit partners (RAP) within the Romanian banks
Source: Authors" projection The results of the research support the fact that within the banking system in Romania, the rotation of the audit partners exists (for 15 analyzed banks for which there could be collected data from the 29 existing ones, about 51.72%) and it is not influenced by the issued audit opinion. Moreover, the period when an audit partner returns to a client is, where the data allow to perform this analysis, longer than 2 years. These aspects present indications that support the independence of the audit firms and the auditors who carry out audit engagements for the banking institutions in Romania.
In respect of the bank country of origin, there are no indications to support the connection between a particular country and the rotation practices of the firm or the audit partner.

Conclusions
External audit is a control and assurance mechanism, which gives interest-holders a reasonable assurance that the financial statements are in line with the reality and they are prepared in accordance with the generally accepted accounting principles. In order for the financial audit to play its extremely important role, the independence of the auditor and the audit firm is a fundamental element which must not be lacking or compromised. The professional literature recognizes the importance of independence in carrying out the audit engagement, without existing contradictory studies regarding its role. Alternatively, the international studies draw attention to the elements that may be threats to the independence of auditors and which are often encountered in practice, in performing the audit engagements. Among these threats there are the provision of non-audit services approved by the regulations in force along with the audit services, the creation of a familiarity between the auditor and the client as a result of a long-term collaboration, the creation of a financial dependence of the audit company on the client which may lead to the latter exercising an influence on the results of the audit engagement, etc. Although the findings regarding the impact of these threats on the auditor independence may be contradictory, the risk exists and should not be ignored. A method to protect independence and to diminish risks and threats towards it, consists in the rotation within the audit engagements. In this case also, the professional studies reveal various opinions regarding the advantages and disadvantages rotation can generate. First of all, due to the knowledge accumulated on a client or the specialization of the auditor in a specific field or industry, its change can be a disadvantage. The rotation policy may also increase competition between the audit firms that may give in to the client to obtain the audit engagement, or may give power to an audit client that wishes to change its auditor which provided an audit opinion that was not in accordance with the client's requirements and expectations. On the other hand, the lack of rotation can cause the auditor to reduce the degree of professional skepticism as a result of being familiar with the client's activity and with the client itself, and can allow the auditor to be biased or influenced by the client. Other issues regarding the rotation within the audit engagements and for which there is no common opinion into the literature refer to the typology of change. Thus, some authors advocate for the rotation of audit partners, while other studies highlight the benefits of the audit firm rotation. There are also researches related to the advantages and disadvantages of the voluntary rotation or of the one imposed by the national regulations.
Thus, we can see that although the concept of independence in carrying out the audit engagements is quite clear and easy to understand, the way in which independence can be maintained in practice generates debates, especially in respect of the rotation of the audit firm or its audit partners.
In Romania, the legal regulations in force do not envisage the mandatory rotation of the partners or the audit firm after a certain period. On the other hand, in our opinion, after analyzing the existing practices mentioned in the professional literature, in order to ensure and maintain independence, the following aspects must be considered: the rotation of the audit firm should be performed after maximum 10 years and that of the audit partner after maximum 5 years; the return of an audit partner to a client should only be achieved after a period of pause of at least 2 years; the rotation of the audit firm and audit partner should be legally regulated; rules should be imposed to prohibit the client from changing the auditor as a result of the audit opinion he or she has expressed.
Our research aims to analyze the rotation policy adopted by the institutions within the Romanian banking system. Although the rotation of partners and audit firms is not compulsory, we believe that such practices indicate a concern of the clients and audit firms in maintaining independence and thus, maintaining the quality of audit engagements.
There are currently 29 banking institutions on the Romanian market. The audit reports for 12 consecutive years for each of the 29 banks were collected, analyzed and centralized manually (where available). The information was collected from the official websites of the banks and from the online environment. Out of the 29 banks analyzed, 7 do not present data on their audit engagements. For the remaining of the 22 banks there were found audit reports but not for all 12 years, and only for certain periods the data were present for consecutive years.
The research was divided into two parts. In the first part, the rotation of the audit firms was analyzed in correlation with the issued audit opinion. The results indicate that only 4 banks (13.79%) display data on the audit firm allowing the verification of the audit firm rotation after a period of maximum 10 years; 13 banks (44.83%) have indications of the rotation of the audit firm (there is an obvious change of the audit firm), but without being able to determine the time period after which the rotation takes place; and for 5 banks (17.24%) there are not enough data to confirm or deny the existence of the rotation of the firm conducting the audit engagement. For the remaining of 7 banking institutions (24.14%), the data needed to perform an analysis could not be collected. The second part of the research considers the rotation of the audit partner in the engagements carried out by the same audit firm, also pursuing the policy of returning of an audit partner after a certain period and the influence of the issued audit opinion. The results indicate the fact that 8 banks (27.59%) make a change of the main audit partner after a period of at most 5 consecutive years, 7 banks (24.14%) have the rotation partner's audit indications (but the lack of data for the preceding and following years makes impossible the real estimation of the rotation period), 5 banks (17.24%) keep an audit partner in office for a period longer than 5 years, and for 2 banks (6.90%) it is not possible either to confirm or to deny the existence of a change of the audit partner.
We mention that in our opinion, the bank country of origin and the audit opinion did not influence the rotation of the partner or the audit firm. We also state that no evidence was found for the data analyzed to indicate the return of an audit partner after a period shorter than two years.
In conclusion, we can say that, although the Romanian regulations do not have rules regarding the need for the rotation of the firm or the audit partner within the audit engagements, the banks operating within the Romanian banking system adopt international practices and in more than 50% they implement their own rotation policy. Although the rotation of the firm or the audit partner may take place at a time interval longer or shorter than the accepted limit or the one practiced in other countries, we consider that simply adopting the rules for changing auditors during audit engagements is clear evidence of the banking institutions and audit firms" concern to maintain independence.
The main limitation of the current research is the lack of a complete data series, over an extended period and for a higher number of banking institutions. Another limitation is the manual processing of data. As future research directions, the analysis will be extended over a longer period of time and the elements which define the financial dependence of the auditor on a client will be included in the evaluation of the non-audit services.
We consider that the present research can be useful to the interest-holders within the banking system, which count on the audited financial statements, the quality of the audit and the independence of the auditors, as well as to the regulatory bodies in Romania, which may consider establishing rules to envisage the rotation of partners and audit firms after a certain period of time, in order to promote and maintain independence in carrying out the audit engagements.