An Assessment of the Provisions of Regulation (EU) No 537/2014 on Non-audit Services and Audit Firm Tenure: Evidence from Spain

On May 27, 2014, Regulation (EU) No 537/2014 was published in the O ﬃ cial Journal of the European Union. Aiming to enhance audit quality, the new regulation establishes important limitations to the selling of non-audit services by the audit ﬁ rm to audit clients and a maximum tenure of ten years with the audit ﬁ rm. However, it should be noted that the extant research has not consistently supported that non-audit services or long tenures impair the quality of audits. This research studies whether these provisions have been empirically associated with reduced audit quality for Spain. Because of its low litigation risk, the potentially negative impact of both non-audit services and tenure on audit quality should be clearly observed in the Spanish audit market. Nevertheless, we do not ﬁ nd signi ﬁ cantly lower levels of audit quality associated with either non-audit services or long audit tenures. However, these results are conditional on the validity of using abnormal accruals to measure audit quality.


Introduction
The 2010 Green Paper on Audit Policy issued by the European Commission (hereafter "the Green Paper") showed serious concern for the independence of external auditors and explicitly encouraged further research on the issue. Only four years after the approval of the Directive on Statutory Audit (2006/43/EC) (hereafter "the 2006 EU Directive"), the European Commission openly questioned the sufficiency of the current regulatory framework to adequately guarantee the independence of auditors. According to the Green Paper, the provision of non-audit services (NAS) by the audit firm to audit clients 1 and the familiarity between auditors and clients as a result of long audit firm tenures would be the main threats to the effective independence of external auditors. As a result of this concern, Regulation (EU) No 537/2014 of the European Parliament and of the Council on specific requirements regarding statutory audit of public-interest entities (hereafter "2014 EU Regulation") has enforced serious limitations to both the types of NAS to be provided to audit clients and the total amount of fees to be charged for these services. Moreover, it has also established a maximum tenure of ten years with the audit firm as the general rule.
However, it should be noted that the concern about the negative effects of NAS and long audit firm tenures on audit quality suggested in the Green Paper has not been consistently supported by the available evidence (e.g., DeFond, Raghunandan, & Subramanyam, 2002;Callaghan, Parkash, & Singhal, 2009 for NAS;Johnson, Khurana, & Reynolds, 2002;Myers, Myers, & Omer, 2003 for tenure). Moreover, in the specific case of firm tenure, since the available evidence does not generally support that long tenures impair audit quality and, given that most of this evidence was obtained before the mandatory rotation of lead audit partners was established, once partner rotation is already mandatory in the EU, the potentially negative implications of long firm tenures on audit quality should be less serious.
In this research, we investigate the impact of both the provision of NAS to audit clients and long audit firm tenures on audit quality. With regard to NAS, we study the total amount of NAS fees charged by the audit firm to its audit clients and the types of NAS provided to these clients. As the 2014 EU Regulation limits the fees for NAS to 70% of audit fees, we investigate whether there is a loss of audit quality when this condition is not met. Next, we classify NAS by nature into audit-related services, tax-related services, and other services, and investigate whether the provision of each specific type of NAS is associated with lower audit quality. Following the same approach, since the 2014 EU Regulation establishes a maximum tenure of ten years with the audit firm, we also investigate whether there is actually a loss of audit quality in audit engagements lasting over ten years. As in prior research, we use discretionary accruals to measure audit quality (e.g., Carey & Simnett, 2006;Myers et al., 2003). The empirical analysis examines a sample of Spanish public companies for the period between 2005 and 2013.
Our motivation for performing this study lies in the controversy surrounding the 2014 EU Regulation. Both the severe limitations to the provision of NAS and the mandatory rotation of audit firms will undoubtedly impact the dynamics of national audit markets in the EU. In countries such as Spain, with a current average audit firm tenure of ten years, the new regulation will necessarily lead to much shorter audit engagements, thus involving serious implications for audit firms. Additionally, after the enforcement of the new regulation, for some audit firms it might be more profitable to give up the auditing of some clients in order to keep the provision of NAS for these clients. Therefore, since the 2014 EU Regulation is expected to significantly affect the configuration of the audit sector in the EU, the next logical step would be to study whether this regulation will actually increase audit quality. Given that the risk of litigation faced by the auditor plays an important role to guarantee audit quality (e.g., Blay, 2005;Khurana & Raman, 2004;Reynolds & Francis, 2000), the Spanish audit market provides an interesting setting for this study. If NAS and/or long audit tenures constitute serious threats to auditor independence, the effects should be more clearly observed in low litigation risk countries such as Spain, where auditors face weaker incentives to maintain independence. We therefore extend prior research on the implications of NAS and audit firm tenure on audit quality by studying the most controversial provisions of the 2014 EU Regulation. While the available evidence on these issues provides relevant insights regarding the rationale behind the new regulation, this research aims to contribute by specifically addressing whether the new regulation could lead to an increase in the quality of audits.
We do not find significantly lower levels of audit quality when audit firm tenure is longer than ten years, as well as when fees for NAS charged to audit clients represent > 70% of the audit fees, or when the audit firm provides the types of NAS more clearly restricted by the 2014 EU Regulation. These results are robust to various checks. Therefore, the evidence reported here does not provide support for the limitations to the audit activity established by the 2014 EU Regulation, and it offers some arguments to the audit profession which has generally opposed the new regulation, given the expected negative impact of these measures on the income statements of audit firms.
The remainder of the article is organized as follows. Section 2 provides a summary of the international regulation of the auditorclient relationship. Section 3 reviews the literature on the impact of NAS and audit firm tenure on audit quality and develops the hypotheses to be tested. In Section 4 we define our model and describe the dataset. We discuss results in Section 5, while in the last section we draw the conclusions and implications of this research.

Regulation of the auditor-client relationship in the EU
As a result of Enron and other financial scandals at the beginning of the century, regulators and policy makers became particularly concerned about the quality of accounting information released by companies. Given the key role played by auditors in guaranteeing the quality of financial statements, the Sarbanes-Oxley Act (SOX) required the Government Accounting Office to carry out a study on the potential effects of imposing the mandatory rotation of audit firms on auditor independence (GAO, 2003). The study found no negative effects of long tenures on the quality of financial reports and thus it did not recommend mandatory rotation. However, the regulator eventually provided the mandatory rotation, although only of lead audit partners, every five years. Moreover, SOX significantly restricted the types of NAS that audit firms could provide to their audit clients. 2 With the same aim as the SOX Act in the US, the 2006 EU Directive also enacted the rotation of key audit partners after a maximum of seven years. Similar to SOX, it did not impose the mandatory rotation of audit firms. Moreover, the regulation of NAS provided by audit firms to audit clients was fairly general. Hence, according to Article 22, audit services should not be provided in cases where "an objective, reasonable and informed third party would conclude that the statutory auditor's or audit firm's independence is compromised." According to the Green Paper, Article 22 had been implemented in a very divergent manner across the EU. For example, while in France there was a total ban concerning the provision of NAS to audit clients, as well as strong restrictions on the possibility for the members of the network of the auditor to provide services to the members of the group of the audited entity, in many other member states rules were much less restrictive.
The Green Paper explicitly put into question the sufficiency of the regulatory framework established by the 2006 EU Directive to adequately guarantee auditor independence. As a result of this concern, the 2014 EU Regulation limits the amount of fees to be