Abstract
The growing complexity of the global business environment is leading innovative firms to demand further external counselling to better handle rapid change and increasing uncertainty. Likewise, the implementation of new information systems is usually carried out with the help of external advisors. This paper examines whether the provision of consulting services undermines audit reporting quality by testing for an association between advisory services and audit reporting. A cross-sectional logistic regression is estimated to test the relationship between consulting fees and the audit outcome. The evidence suggests that there is no statistically significant association between non-audit fees and audit outcome. This finding is consistent with the idea that audit reporting quality is not impaired by the provision of consulting services.
Acknowledgements
The authors are grateful to the University of Valencia, Spain for the financial support from Research Project UV-AE-24161 and to the Ministry of Science and Technology of Spain for the financial support from Research Project ECO2009-14457-C04-04.
Notes
This Act was passed in response to a number of financial scandals that caused great concern in the international business environment (Enron, Woldcom, Parmalat…) and it attempted to recover economic agents' trust on financial markets. The act also included specific measures such as the requirement for audit team heads to take turns; the setting up of an auditing committee for listed firms; and the extension of the auditors' incompatibilities based on potential conflicts of interest.
ICAC: acronym for Instituto de Contabilidad y Auditoría de Cuentas.
United States Securities and Exchange Commission.
It is argued that these benefits occur because of the lower costs associated with the joint provision of services (Abdel-khalik, Citation1990; Palmrose, Citation1986; Simunic, Citation1984).
CNMV: Comisión Nacional del Mercado de Valores. In english, National Securities Exchange Commission.
Most variables analysed did not accomplish the assumptions required for parametric tests such as normality.
Indeed, the experimental variable is significant at 10% level (p-value = 0.085).