Abstract:
This thesis examines the effect on auditor independence of providing non-audit services in the Norwegian audit market. I generally follow Hay, Knechel & Li (2006) and use three ways to investigate potential loss of auditor independence. The analyses are based on a sample of Norwegian public firms from 2008 to 2010. I find that there is a typical positive relationship between audit fees and non-audit fees when a single OLS regression model is used. Also, there is no significant relationship between the provision of non-audit services and either the frequency with which auditors issue qualified or modified opinions or audit tenure. I use additional tests to examine if results are sensitive to different model specifications and alternative explanations. The overall results of robustness tests are very similar to previous results except for three new findings. First, the positive relationship between audit fees and non-audit fees disappears when a two-stage least squares model is initially implemented; however, the relationship becomes positive again in replicated models which include firms with zero non-audit fees. Second, the positive relationship between audit and non-audit fees in the replicated models occurs because of the implication that ‘joint determination of audit fees and non-audit fees’ does not hold in Norway. Third, some evidence suggests the existence of knowledge spillovers in the Norwegian audit market when both audit and non-audit services are provided by an auditor and the level of non-audit fees is relatively high. In summary, my findings suggest non-audit services may lead to loss of independence only in appearance. The overall results fail to find any evidence for loss of independence of mind as a result of providing non-audit services in the Norwegian audit market.