AUDIT FEES AND AUDITOR CHANGE; AN INVESTIGATION OF THE PERSISTENCE OF FEE REDUCTION BY TYPE OF CHANGE

Despite concern expressed over 'low-balling' or competitive price reduction in the market for audit services, the only empirical investigation of the relationship between initial audit fees, fee recovery and auditor change which has been carried out is by Simon and Francis (1988) using US cross-sectional data, although Pong and Whittington (1994, forthcoming) examine the impact on first year fees. We investigate the impact of auditor change on cross-sectional UK data for 1991 with the aim of discovering whether there is any evidence of price cutting and subsequent price recovery in a European, rather than a US context. We find that there is evidence of such a phenomenon, and investigate further to see whether this effect is robust with respect to the definition used of 'auditor change', and whether the effect varies according to whether the old auditor was a 'Big Six' or 'non-Big Six' firm. We find that the largest reductions (controlling for auditor size) are experienced by firms changing to 'Big Six' firms. We further show that involuntary changes (by auditor acquisition) are associated with positive (although statistically insignificant) increases in fees, whereasvoluntary changesare associated with negative (andsignificant) feereductions. These effects are shown to be robust with respect to model specification. This suggests that fee reductions cannot be ascribed to economies of scale or scope, but is consistent with a 'low-balling' model of behaviour.     

AUDIT PRICING, AUDITOR CHANGES, AND THE WINNER'S CURSE

This paper presents a two-period model of the audit market. In the first period, all auditors have symmetric information and adopt identical bidding strategies. In the process of performing the audit, the incumbent auditor learns the actual costs, thereby becoming informationally advantaged in the second period. In the model presented, unlike earlier ones found in the literature, audit costs include both a component common to all potential auditors and a private component that varies across auditors. The common component of auditor costs gives rise to a ‘winner's curse’ scenario. A winner's curse is said to exist because a non-incumbent bidder who does not take into account the superior information of the incumbent would be expected to generate a loss from winning the audit engagement. The adjustment of bids by sophisticated auditors to compensate for the winner's curse is shown to play a significant role in determining the degree of low-balling (first-period price cuts) and auditor turnover. In the model, low-balling is not associated with loss of audit quality. Additionally, it is shown that it is in the interest of the client to structure audit selection in a manner that gives rise to low-balling.     

EXTERNAL AUDITOR EVALUATION OF THE INTERNAL AUDIT FUNCTION

This study examines the relative importance of the factors used by external auditors when valuating an internal audit function. The study also examines the consistency of external auditors in making evaluations of an internal audit function. The factors used are based on Statement of Auditing Practice AUP 2 “Using the Work of an Internal Auditor” and a similar study conducted in the U.S.A. The two factors which this study finds to be the most significant are ‘technical competence’ and ‘due professional care’. The study finds a high degree of consensus across the respondents with respect to the evaluation of the internal audit function, and a high degree of insight and stability in their judgements.     

AUDITOR INDEPENDENCE,AUDIT QUALIFICATIONS AND THE PROVISION OF NON-AUDIT SERVICES: A NOTE

This paper presents the results of a study designed to investigate the question of whether there is the potential for an appearance of auditor independence impairment when higher levels of non-audit services are provided to audit clients. This question was investigated by analysing the audit reports for a sample of publicly listed companies over the period 1980 to 1989. The results of the analysis indicate that the auditors of companies not receiving an audit qualification of any type over the period derived a significantly higher proportion of their remuneration from non-audit services fees than the auditors of companies receiving at least one audit qualification. This result is driven by the non-Top 100 companies in the research sample. These findings could suggest that auditors are less likely to qualify a given company's financial statements when higher levels of non-audit services fees are derived. This could convey the view that the independence of the auditors of these companies may have been compromised.     

AUDITOR OBJECTIVITY: THE EFFECT OF PRIOR INVOLVEMENT IN AUDIT PROGRAMME DESIGN

This study tests whether the involvement of independent auditors in the design of an audit programme affects their subsequent audit work relating to that audit area. Two groups of subjects played roles in designing partial audit programmes for a particular account. They also evaluated partial audit programmes for another account designed by someone else. A third group merely evaluated partial audit programmes for both accounts. All subjects were informed that both audit programmes were implemented and of potential problems that could be attributed to either of the two accounts. Subjects were asked to allocate 15 hours to investigate either or both of the accounts in question. Our hypothesis was that auditors who had a role in designing a partial audit programme would allocate less audit effort to that account than auditors who had no role in the programme design. Comparisons of the hours allocated provided some support for this hypothesis.     

DETERMINANTS OF AUDIT FEES FOR QUOTED UK COMPANIES

This study reports further evidence as to the determinants of the audit fees paid by quoted companies in the UK. It outlines a framework based on the findings from semi-structured interviews with partners in four large audit firms and the results of previous research, and tests this framework by means of multivariate analysis using 1987 data for a large sample of quoted UK companies. A model explaining 87 per cent of the variation in audit fees is constructed. The principal explanatory variables are found to be auditee size, return on shareholders equity, the number of subsidiaries, the lag between the year end and the date of the audit report, the size of the auditor, a measure of auditee diversification, the ownership structure of the auditee, and whether the auditor was based in London, with the last three being new variables introduced in this study.     

THE DETERMINANTS OF AUDIT FEES: SOME EMPIRICAL MODELS

This paper proposes a model of audit fee determination based upon size and other characteristics of the auditee and the auditor, and upon whether there has been a recent change of auditor (which might lead to a low-balling effect). A novel feature of the model is that auditee size is measured in two dimensions, sales and assets, and is allowed to have a quadratic relationship with audit fee. The model was estimated for a sample of large listed UK companies for the years 1981 to 1988. The results support the low-balling hypothesis and also provide insights into the distinctive role of the Big 8 firms in the audit market.     

AUDITOR-GOVERNMENT ASSOCIATIONS AND AUDITOR INDEPENDENCE IN CHINA

Chinese CPA firms were established and initially owned by government bodies. This situation has caused much concern regarding auditor independence in China. A program to disaffiliate CPA firms from their sponsoring body began in 1997, and the purpose of this paper is to investigate whether auditor independence has subsequently improved. We use the number and percentage of non-standard opinions in audit reports as a measure of auditor independence. We found that the number and percentage of non-standard opinions has increased dramatically since 1997. After considering other possible explanations, we attribute the cause for such a change to the disaffiliation program. These results are consistent with the notion that disaffiliated CPA firms face a higher degree of market risk than affiliated firms, and therefore act more independently. Our results thus provide support and justification for the disaffiliation program.     

AUDITOR CHANGE ANNOUNCEMENTS AND DISPERSION OF INVESTOR EXPECTATIONS

The effect of auditor change announcements on the dispersion of investor expectations is investigated by using two approaches. First, the consensus effect (Holthausen and Verrecchia, 1990) is measured by examining unanticipated trading volume change. We show that auditor change announcements provide information to the market, and that the consensus effect dominates the informedness effect. Second, the reduction in information asymmetry (Glosten and Milgrom, 1985) due to auditor change announcements is demonstrated by a reduction in proportional bid-ask spread, which is not driven by increased trading volume. Thus, auditor change announcements reduce dispersion of investor beliefs in that they are both information asymmetry-reducing and consensus-increasing. Finally, we show that the type of auditor change (Big-Eight to Non-Big-Eight, Non-Big-Eight to Big-Eight, or Within-Class) has no impact on the dispersion-reducing effect of auditor change announcements.     

DETERMINANTS OF ACCOUNTING POLICY CHOICE IN THE AUSTRALIAN PROPERTY INDUSTRY: A PORTFOLIO APPROACH

The objective of this research is to explain accounting policy choice in the Australian property industry by reference to costly contracting theory. This paper makes three departures from the predominant literature. First, instead of focusing on a single policy choice, it focuses on a portfolio of 15 choices. Second, instead of measuring each policy choice dichotomously, it measures the dollar effect of each policy choice. Finally, the research focuses on the impact of policy choices on owners' equity as well as income. The results confirm the importance of compensation arrangements as a determinant of accounting policy choice but fail to provide evidence that debt constraints or political costs are significant determinants of accounting policy choice. These results suggest that including property firms in a more heterogeneous sample may reduce the power of tests on the debt constraints and political costs hypotheses. More generally (when compared to previous research) the results suggest that the determinants of policy choice are likely to be industry specific.     

AUDITOR PREFERENCES FOR LIABILITY LIMITATION

Debate over statutorily limiting auditor civil liability has implicitly assumed auditors are homogeneous in their preferences for capping liability. This study examines the preferences of auditors for limiting auditor liability and investigates reasons for the preferences. The study uses an Australian setting in which there has been a persistent debate for a decade or more over regulatory intervention in this area. The study provides a background to the debate over this issue and addresses the effects of two factors suggested by the extant literature, namely auditor size and the business risk of an auditor's client portfolio. These factors are argued to affect the expected costs of litigation facing auditors and therefore their preferences on capping liability. Using the submissions by audit firms on an Australian Companies and Securities Law Review Committee Discussion Paper on limiting auditor liability, the study finds larger audit firms that have greater capacities to lobby and greater expected costs of litigation from unlimited liability than smaller firms, dominate the respondents on the Paper and tend to be more supportive of liability limitation than smaller audit firms. Within the array of possible methods of capping liability canvassed by the Discussion Paper, the study documents evidence of diversity in preferences among audit firms. Larger audit firm size is associated with a preference for a group of methods that provides such firms with opportunities to benefit from the capping at the expense of the smaller audit firms. The method most preferred by the larger audit firms is the multiple of fee with a prescribed minimum. Perhaps not surprisingly, this is also the preferred method of the professional accounting bodies in Australia. As to the effect of the riskiness of the client portfolio on preferences for methods of limiting liability, the study finds that higher business risk in an auditor's portfolio is associated with a preference for methods that give greater control over their liability exposure. The study has implications for the impact of regulation of capping liability on competition in the audit services market.     

AUDITOR LIABILITY AND THE DEVELOPMENT OF A STRATEGIC EVALUATION OF GOING CONCERN

This paper addresses various aspects of the auditor liability problem in Australia. It identifies, as the fundamental cause of the auditor liability problem, the moral and ethical degradation of the accountancy profession and, as a derivative thereof, its intellectual impoverishment. It indicates a significant imbalance between self-interest and public-interest in the profession. This imbalance is supported with an examination of the resolute efforts which were actioned by powerful members of the profession to control the initiatives which were taken to protect a small minority of members, including themselves, from damage awards. It examines the success which those parties had in bringing about the introduction of an act of parliament to cap damages awarded by a judge and it explores the obnoxious scheme that was enacted under that act to protect wrongdoers from the payment of just, fair, and equitable damages to their victims. The paper indicates that the profession did not do its job properly and that, accordingly, it is not so much the victim as it is the culprit. The paper identifies the evaluation of the going concern basis as being fundamentally flawed and, thereby, as being a basic cause of auditor liability. It suggests the introduction of a multi-disciplinary approach to such evaluations utilizing, amongst other things, strategic management dimensions. It also proposes an order of magnitude expansion of the role of going concern basis evaluations. The paper indicates that the profession is a victim of itself. It concludes with the observation that the accountancy bodies are myopic and that they are not learning organizations. This situation, it states, must change.