This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
This chapter largely deals with the vital point of current knowledge, extensive secondary research and publications related to the topic of the study. It classifies and elaborates on the concepts of introduction and development of external audit fees from a global perspective; prior researches conducted in the Sri Lankan and South Asian context and the identification of factors affecting eternal audit fees. The review is also an effort to understand the determinants of audit fees and hypotheses for an audit fee model for the listed companies in Sri Lanka.
In order to help with a better understanding for this study, it is greatly supported by the previous review of studies which have taken place in 1980s and 1990s, on external audit pricing in a universal perspective.
The purpose of this review is to analyse literature on the audit landscape, thereby to assess the current understanding of the major themes relevant to addressing the research question of this thesis: What are the determinants of audit fees in the Sri Lankan context? There is an abundance of literature on determinants of audit fees, pertaining to what will be classified in this thesis as the audit fee model. However there is only a limited amount of literature available on audit fee research from a Sri Lankan organisation standpoint. Similarly it must be noted that audit fee determinants on listed companies in Sri Lanka are constantly evolving due to major changes in the business environment and how organisations do business.
Development of audit fee in a global perspective
During the last couple of decades the 'dot.com' bubble burst and major corporate collapses took place; the reverberations from the events of 11 September 2001; the introduction of Sarbanes Oxley (SOX) requirements in 2002; the transition in 2005 to International Financial Reporting Standards (IFRS); auditing standards becoming legally enforceable in 2006 and being further 'clarified' in 2010; the 2008 Global Financial Crisis (GFC) causing financial distress on a scale not seen since the 1930s; a post-Global Financial Crisis debt crisis plaguing the United States and the European Union sum up the volatile global expanse. Sri Lanka however may not have been significantly impacted by world events, yet natural disasters like the Tsunami, a raging 30 year ethnic war which brought about unprecedented stock market activity and devaluation of the rupee calls for proper reform and financial regulation. The direction given in the Colombo Stock Exchange Annual report 2011-2012 shows that the monopolised activity witnessed on the Colombo Stock Exchange could have been better monitored and regulated leading to harmonious investments and foreign capital inflow. To this effect recommendations of the code of best practice on governance issued jointly by the Securities Exchange commission and the Institute of Charted Accountants of Sri Lanka have been set and listing rules have been revised in February 2012.
According to the sustainability reporting initiative guidelines in 2007, it is identified that the requirement for a broader information set is clearly demonstrated by the small percentage of market value now represented by physical and financial assets due to the interdependencies in economies, advanced technology and increased demands and consumption levels. The rest, intangible factors are very unlikely being explained in the financial statements. The type of information required to assess historical, current and future prospects is much wider than is provided for by the existing reporting model.
External reporting is now due for a wave of modifications. The International Integrated Reporting Council (IIRC) is developing an integrated reporting framework that will monitor the improvement of external reporting over the next years. It has issued a discussion paper (IIRC, 2011) and intends to issue an exposure draft in 2012. For auditors, these changes have increased the risk and complexity of auditing financial-statements and highlighted the importance of audit quality to market confidence. As the 'contracted regulators' of financial reporting, auditors have inevitably increased audit effort in response to these higher risks. One measure of greater risk is the increasing rate of corporate insolvencies.
To this effect Simunic (1980) primarily developed a positive model in the market for audit in publicly listed companies with respect to the price competition and the study was mainly based on a sample of 397 observations.
The model recognised three basic variables which are,
loss exposure variables represented by the total assets and the complexity in terms of decentralisation (Number of subsidiaries) and diversification
loss sharing variables represented by the ratio of net income to total assets, the loss incurred by the firm as a dummy variable and the opinion of a qualified auditor through a dummy variable
production function variable represented through the auditor tenure
The R squared value which was 57% for the total sample defined a nonlinear relationship between total assets and the audit fees. In order to linearise the relationship between the two variables he then used the transformation of square root of the independent variable, namely total assets.
Secondarily, he classified according to the form of industry which he categorised as utility firms and banking sector. The two categories were treated as dummy variables which labelled as 0 and 1 respectively. R squared was then identified as 42%.
The whole sample was categorised into two on the basis of size of sales which are more and less than $125 for to test the impact of size on audit fees and achieved a R squared value of 28% and 51% correspondingly. The results showed that the variables which calculated for loss exposure are statistically significant. Regardless of identifying the determinants of audit fees he also concluded that there is no dominated pricing which leads big firms to enjoy economies of scale to charge lower fees.
The limitations of this study is that the data has been gathered through questionnaires, a qualitative method which gives rise to measurement errors, question of reliability of collected data and is also a moral hazard to empirical research analysis. The results of the study are also restricted to a certain time period and therefore extra studies have to be conducted with a broader period of time.
With reference to the empirical work of Elliot and Korpi (1978) where he carried out a quantitative study with two different sectors comprised of manufacturing and financial of 60 and 42 companies each in the respective sectors. The data was collected over questionnaires to develop an audit fee model. Degree of complexity, size of sales and total assets were considered in this study as the factors and moreover the square root transformation was places to linearise the relationship with its dependent variable.
The model described 94% and 84% variance of audit fees for manufacturing and financial companies respectively which is treated statistically significant.
The results of this study provide a clear explanation only around two specific sectors and therefore it could be only accurate to explain the reliability of results within these two industries. It will be much precise to apply the same model in a much wider range of companies ranging across multi-disciplined industries where we could generalise the findings to a larger sample.
Briston and Perks (1977) assessed the total audit fees paid by all the 592,243 companies; 3,555 listed companies & 589,688 unlisted companies, ranked by turnover registered in England, Scotland, and Wales to be around £200 million comprising £109.70 million for the listed companies plus £90.0 million for the unlisted companies for 1975 / 1976, and £250 million during 1977. They used the average of the audit fees as a percentage of turnovers (i.e., 0.1 %.) to estimate the total audit fees. Their study indicated that the proportion of audit fees to turnover decreases as the company's size increases e.g., audit fees as a percentage of turnover was 0.1 for the companies with total turnover £2.2 billion, and it decreases to 0.04% for the companies with total turnover £69 billion.
Fanning (1978) also estimated the total audit fees of limited companies in the U.K. including the top 100 quoted companies by market value, 500 other quoted companies, and 300,000 unlisted companies in 1976 / 1977 to be £416 million comprising of £54.42 million for the top 100 companies, £212.85 million for the other 500, and £148.75 million for the 300,000 unlisted companies. Additionally the estimated audit fees for 1978 was about £450 million assuming a continuing trend in fees increases between 15% and 30%. The increasing risks, complexity and costs of financial-statement audits would suggest substantial increases in audit fees, particularly post 2008 period.
These models give an indication that the size of a company in terms of turnover is a critical component that needs to be considered. However it may not provide enough clarity to determine that higher turnovers alone are inversely proportional to the audit fees charged. Further research areas could be performed to analyse the companies across industries and in terms of market capitalisation and other financial ratios.
The study conducted by Taylor and Baker (1981) with only 126 manufacturing companies in the United Kingdom and the same square root transformation was followed as it has enriched the association among the dependent and independent variables. The factor analysis has been conducted as a data reduction method in recognising the most feasible measures of size and complexity and has found that total assets and the number of subsidiaries held under the individual company are the most viable measures. The developed audit fee model has explained 79% of the variation in the audit fee which is treated as highly significant.
Wayman and Bavishi (1983) conducted a cross country study to determine the impact of company size on audit fees for 640 companies across four countries; India, Malaysia, United Kingdom and Australia. The independent variables of the regression model are identified as the size of auditor and the client, financial year end, nature of business (i.e.: multi-nationality) while the audit fees remains as the dependent variable.
On application of the model, it explained variances of 69%, 51%, 88% and 49% respectively in Australia, India, Malaysia and the UK. An inverse relationship was found between the two variables which was same across all countries.
One of the drawbacks of this study is that this is conducted in a quite isolated countries spread in the globe and therefore may not provide a universal view.
Additional areas for research would be to examine a more definite geographical area such as in specified regions (i.e.: European region, South Asian region) and develop the audit fee model.
Chan et al. (1993) assumes as the fact that a high insider ownership lessens agency conflict between manager and shareholders. A preliminary sample of 985 UK publicly listed companies was used in this study. The gathered data were then categorised as big firms and non -big firms. The overall results proved an inverse relationship between audit fees and insider ownership and it was the same with the subsample too which is non-significant for the small firms in the sub-sample.
Using the same argument as Chan et al. (1993) on a sample of Norwegian firms, Firth (1997) discovers no significant relation between insider ownership concentration and audit fees. Piot (2001) finds a non-significant relation between insider ownership and the choice of big audit firm in France.
Niemi (2005) tests the same model of Chan et al.'s (1993) on Finnish firms and finds a non-significant relation between audit fees and the measure of the combined managerial and non-managerial ownership concentration. Then, the author explains these mixed results by the fact that managerial and non-managerial ownership concentration should have opposite effects on audit fees. He found a significant negative relationship between audit fees and management control of the firm and also a positive relationship between audit fees and state control and foreign holding control. The author acknowledges the collection bias in that the countries selected, all have strong "european" ties, and "thus they may not reflect a varied set of international audit fee pricing practices". It is accepted that, although prior literature indicates some harmony on the variables chosen, other variables had also been identified which could further explain audit fees. The evidence is by no means clear as to whether selection of other variables would have shaped any more powerful explanatory model. The results are normally positive and significant, across countries and firms, although the level of applicability speckled between country and firm. The researcher considered this uniform in the model across ranges of firms and countries however may not hold factual for Asia centric countries.
Fan and Wong (2005) conducted a study on audit fees determinants in Asia, where family ownership is high and investors are less secured. They test whether auditors request for an additional premium from their clients when agency clashes are high. The authors find a positive relationship between audit fees and ownership concentration and explain that auditors assume a higher risk to audit those firms. Previous research therefore presents varied proof of the effect of agency conflicts on the supply of audit services and points out the requirement to consider the discrepancy between supervisory and monitoring shareholders ownership.
Hay et al (2006) reviews a large body of audit fees determinants using a meta-analysis and conclude that the results on relation between ownership structure and audit fees are diverse. Previous studies have focused mainly on the impact of insider ownership on audit fees.
Prior researches in the Sri Lankan & South Asian context
Sri Lanka is considered as a suitable focus for the current study because it is a developing country, has largely have met its accounting and auditing standards to International Financial Reporting Standards (IFRS) and International Standards on Auditing (ISA) equivalents, correspondingly, and has accepted the issues associated with Fair Value Accounting (FVA).
In terms of background to the auditing landscape in Sri Lanka the Institute of Chartered Accountants of Sri Lanka (ICASL) sports a membership of over 3,500 members and was founded by the Act No. 23 of Parliament in 1959. It is the only body approved under the Sri Lanka Accounting and Auditing Standards Act No. 15 of 1995 to issue both Sri Lankan Accounting Standards (SLAS) and Sri Lankan Auditing Standards (SLAuS). Under the latter Act, compliance with SLAS is essential for "specified business enterprises", while SLAuS must be applied for audits of units stated in the Act. As noted in the Introduction, ICASL has issued a series of standards which recommend FVA for specific assets and liabilities.
Despite a few high-profile corporate let-downs, which have generated discussions by stakeholders with regard to the auditors' role, there have been no legal cases or other disciplinary actions against auditors. In the Sri Lankan context there had been few failures of several finance companies in the late 80s and early 90s, the Government of Sri Lanka recognised the Presidential Commission on Finance and Banking to inspect the causes of such failures and to commend actions for firming up the country's financial sector arrangements. The Commission stressed the necessity for adoption and implementation of accounting and auditing standards for establishing a high-quality corporate financial reporting system. Furthermore, the Commission suggested better performance of responsibilities by auditors in averting such failures. Even with these measures, as recent as 2003, there was a failure of a savings and development bank in Sri Lanka. Despite such failures, there have been no legal actions or other administrative actions taken against auditors according to the report on the "Observance of standards & codes" (ROSC) Sri Lanka 2004.
Prior research done on the Sri Lankan context was not available, however from a Commonwealth and South Asian context particularly, the literature includes researches set in Bangladesh (Karim & Moizer, 1996), Singapore (Low et al, 1990), Singapore and Malaysia (Poitras et al, 1995), Singapore, Malaysia and Hong Kong (Simon et al, 1992), India (Dugar et al, 1995) and Pakistan (Ahmed & Goyal, 2005).
Poitras et al (1995) selected a sample of 50 companies by largest capitalization as at December 31, 1991 listed on the Stock Exchange of Singapore (SES). These companies had operations mainly in Singapore and Malaysia. Based mainly on the Simunic (1980) model, with adjustments for local conditions and the introduction of few more variables such as audit fees/total assets, and found find mixed results in their study. However the Pearson's correlation analysis suggests audit fee, as the dependent variable is correlated to audit size and tenure. Singapore exhibits the presence of Big six premium but the results are mixed with an adjusted R square of the empirical model of 92%, with major volatility in audit pricing being observed at due mostly at industry level and on time series basis, the intercept indicates both negative and positive.
The research by Ahmed & Goyal (2005) illuminates a weakly-researched issue as to how auditors charge for their services by empirically investigating audit fee determinants in three developing economies in South Asia. Using 1998 data from 118 Bangladeshi, 219 Indian and 229 Pakistani firms, the results show that size of reporting entity, multinational relationship and size of audit firm are the three most vital determinants across the three countries. They find no substantial relationship between audit fee and a firm's financial state and complexity of the auditee. Moreover, no noteworthy relations were found between audit fees and the firm's financial condition and auditee complexity. The results of this study provide useful visions into the role of contracting cost variables and auditors' billing practices in relative and developing economies in general, and precisely in South Asia.
Naser & Nuseibih (2007) studied the structure of audit fees in Jordan using traditional measures of determination such as size and complexity. In line with prior research, these authors find similarity in their findings with core determinants being both size and complexity in Jordan. In much the same geography, Al-Shammari et al (2008) studied the determinants of audit fees in Kuwait. This would provide clarity on the determinants that impact audit fees from a Middle Eastern perspective. This study is in line with the audit variables that are used and size and complexity are critically analyzed for all intents and purposes.
Simon et al (1986) found similar results similar to previous research in India. Again, Size and Complexity were key determinants of audit fees. They determined that audit fees in India are determined primarily by audit size, complexity and risk. They also suggest that bigger firms with dispersed shareholding are increasingly demanding Big 4 auditors - thus perhaps mitigating agency conflicts - as the majority of "large" shareholders are still "families" who have controlling interests in the firms. They also suggest that while audit premia is not evident in their research, auditors may charge higher fees than Non-Big 4 firms through setting lower qualification levels. This suggests that firms with lower agency issues deploy the services of Big 4 auditors more than firms with a high degree of agency problems. Later research by Dugar et al (1995) is of relevance. This research is consistent with the concept of including the analysis of the big 4 audit firms as a variable to understand its relationship to determine the audit fee model.
Identification of factors affecting the level of audit fees
Hobgood and Sciarrino (1972) predicted factors through replies for a survey carried out in 155 Canadian companies. These were recognised as, size and capability of the audit firm staff, scope of the audit, size of the client and comparison of audit fees salaried by other companies comparable in size and industry.
Morgan (1973 and 1974) correspondingly studied on the objective and subjective factors which impacts on determining the audit fees and he established that time is the main impartial component whereas value of the service to the client and difficulty of engagement as the subjective elements. Zweig (1978) stated that the value of the service be the basic element of the fee determination.
Flesher and Loroosh (1980) recognised nine factors in determining the audit fees through a survey of randomly selected 232 CPAs which are;
â€¢ Time spent on the task
â€¢ Cost of the service rendered
â€¢ Ability of client to pay
â€¢ Value of the service rendered
â€¢ Customary fees
â€¢ Legal limitations
â€¢ Established client (status)
â€¢ Fixed fees
â€¢ Urgency for work to be completed
Finalising the survey, time, cost and the value of the service to the client were known as the most critical factors and others were treated as less important by the respondents.
One unique feature in the case of Bangladesh in 2007 is that there are formally no Big Four firms with an established presence in the country, as in India and Pakistan, even though the population of Bangladesh is around 147.4 million. Consequently, Karim & Moizer (1996) in their research set in Bangladesh, used as a proxy for Big Four firms, local audit forms with links to international audit firms. In their study of determinants of audit fees in Bangladesh, Karim and Moizer tried to determine the factors towards explaining the audit services market in the Country and state that "audit is a buyer's market". Their study suggests a limited Big Six premium effect; however they do state that audit size, audit risk and audit complexity form the foundations for audit price. This research focuses on the big 4 audit firms in the local context to understand their relationship to the model.
Finally a conclusion will be drawn identifying the key points addressed, the conflicts in the literature and the areas this study will address. The main objective would be to understand the determinants for external audit in the listed companies and thereby to understand the development of an ideal audit fee model.
The exploratory nature and limited scope of this research points to a range of related further research. It would be useful to build on this study by widening the range of research into audit fee models from this research, and develop more qualitative determinants, as appropriate. This would create a balanced approach that would perfect the outcome of determining audit fees.
Having regard to the previous researches done on designing the audit fee model this research attempts to demonstrate relationships between variables that would represent risk, size and complexity and related audit fees in the financial statements of the listed companies in the Colombo Stock Exchange.