Audit committee effectiveness and non-audit service fees: Evidence from UK family firms

Abstract

This study examines whether the presence of an ineffective audit committee at family and non-family firms can influence the firms’ non-audit service purchases from the incumbent auditor. Using a sample of 1736 observations of UK-listed companies from 2005 to 2013, we find a significant positive association between ineffective audit committees and non-audit service fees. This association is more pronounced for family than for non-family firms, suggesting that in the UK, family firms with ineffective audit committees tend to buy more non-audit services from their incumbent auditors than non-family firms. The results also show that family firms with ineffective audit committees pay higher non-audit service fees when their family members own shares or hold board positions, indicating that both types of involvement lead to larger non-audit service expenditures.

Introduction

Family firms are the backbone of the world economy (Trotman & Trotman, 2010). According to the US Census Bureau, about 90 percent of US firms are family-owned or controlled, and they account for half of the nation's employment and gross national product (GNP). In the UK, two-thirds of firms, or about 4.8 million businesses, are family-owned, and more than 16,000 of these firms are medium and large businesses (UK Institute for Family Business (IFB), 2017). They generate over a quarter of the UK gross domestic product (GDP), and many are recognized worldwide for their innovative and long term outlook. Despite focusing more on the long term than non-family firms, family firms also devote their energies to the present because they operate within a complex and unique work setting shaped by the overlap of family and business issues.

Over the last decade, there has been widespread global interest in the study of listed family firms and the differences between family and non-family firms in terms of corporate financial performance, quality of financial reporting, and corporate disclosure (e.g., Anderson and Reeb, 2003, Prencipe and Bar-Yosef, 2011, Poutziouris et al., 2015). Recently, emerging researchers have questioned the relationship between firm type (family vs. non-family) and non-audit service (NAS) fees. For instance, Kang (2017) found that US family firms tend to purchase more NAS from their incumbent auditors than their non-family counterparts. However, researchers have not considered the role of corporate governance, specifically the audit committee (AC), in evaluating the NAS fees that family firms pay to their external auditor.

In the wake of recent financial scandals, regulators have been strengthening the AC function, including independence, composition, expertise, and disclosure of activities (e.g., SOX, 2002, Smith, 2003). The UK Corporate Governance Code (2014, p. 5) states that “the board and its committees should have their appropriate balance of skills, experience, independence and knowledge of the company to enable them to discharge their respective duties and responsibilities effectively.” Whereas the monitoring role of effective boards of directors depends on their structure and organization (Peasnell, Pope, & Young, 2005), we contend that the monitoring role of ACs depends on their expertise, composition, and activity.

In the UK, the AC is generally perceived as the cornerstone of the process that oversees the integrity of financial statements and develops and implements the NAS policy (UK Corporate Governance Code, 2014). Empirical studies document mixed evidence on the association between audit committee effectiveness (ACE) and NAS fees. For instance, Abbott, Parker, Peters, & Raghunandan (2003) found that independent and active AC members are negatively associated with the magnitude of NAS fees. Similarly, Abbott, Parker, & Peters (2011) report a negative relationship between ACE and auditor-provided NAS, specifically in settings where NAS fee disclosures are mandated. However, other studies found either a non-significant or a positive association between AC characteristics and NAS fees (Zaman, Hudaib, & Haniffa, 2011). Notwithstanding this information, the literature has failed to examine the impact of ineffective ACs on the NAS fees of family firms.

Firms have distinct governance needs based on firm-specific and environmental conditions (Leung, Richardson, & Jaggi, 2014). Family firms can differ from non-family firms in their governance structures (Songini & Gnan, 2015) and, therefore, in the effectiveness of their AC. Ineffective ACs can lead to a host of problems. For example, “auditors assess family firms with weak ACs to have the highest fraud risk and to be the least desirable audit clients” (Krishnan & Peytcheva, 2019, p. 1). If auditors perceive increased fraud risk with a family firm (in line with the entrenchment theory), they might seek a risk premium or even reject the engagement to avoid the risk exposure (Johnstone & Bedard, 2004). Accordingly, family firms should prefer effective ACs, which would decrease auditors’ perceptions of risk and consequently reduce the firm’s economic and reputational costs.

Relying on agency theory (Jensen & Meckling, 1976) and resource dependence theory (Beasley, Carcello, Hermanson, & Neal, 2009), and using insights from prior literature (Abbott et al., 2003, Abbott et al., 2011, Ye et al., 2011, Zaman et al., 2011, Kang, 2017), this paper investigates the impact of ineffective ACs on NAS fees paid by UK-listed family and non-family firms. Specifically, we examine the interaction of ineffective ACs and family control and involvement on NAS fees. The sample is drawn from the FTSE (Financial Times Stock Exchange) 350 companies listed on the London Stock Exchange between 2005 and 2013.

Firms that involve family members, founders, or descendants are referred to as family firms. These members are the principal owners of the business (majority stake), and they hold top management positions ((Chief Executive Officer (CEO) or Chief Financial Officer (CFO)) and/or sit on the board of directors (Anderson and Reeb, 2003, Ali et al., 2007). Family firms have been found to have better performance, better management control, lower cost of debt (Anderson & Reeb, 2003), and lower earnings management (Al-Okaily, BenYoussef, & Chahine, 2020) than non-family firms. However, compared to non-family firms, they are more likely to exhibit material weakness in their internal control and commit fraud (Chen, Chen, Cheng, & Shevlin, 2010). Also, they may have a longer relationship with their auditor (Khalil, Cohen, & Trompeter, 2011), which can threaten auditor independence (Dobler, 2014, Krishnan and Peytcheva, 2019). They also tend to recruit lower-quality auditors (non-Big 4)¹ and incur lower audit fees (Ho & Kang, 2013).

Family firms have specific agency problems. Compared to non-family firms, they deal with fewer Type I agency problems due to the “alignment effect”², which occurs when family members are both owners and managers (Ali et al., 2007). However, they face significant Type II agency problems because of the conflict between large and small shareholders.³ In this problem, managers/owners act in their own interest rather than in the interest of small shareholders. To minimize agency issues, companies are expected to have strong corporate governance structures, especially when they are family firms. The AC, a central pillar of effective corporate governance, is actively involved in several audit-related activities. Specifically, it plays a vital role in overseeing the integrity of financial statements, reviewing the auditor’s independence and objectivity, and developing and implementing the NAS policy (UK Corporate Governance Code, 2014).

This study is the first to examine whether ACE can affect the purchase level of NAS by family vs. non-family firms. In doing so, it responds to Trotman and Trotman’s (2010) call to compare NAS purchases by family and non-family firms and Kang’s (2017) suggestion to investigate the impact of corporate governance factors on the latter association. It also fulfills Tepalagul and Lin’s (2015) invitation to study NAS in countries with low litigation risk, unlike the US. The UK was selected, as it is known to have a less litigious setting than the US (Wu, Hsu, & Haslam, 2016).

Our study contributes to three streams of research. The first stream analyzes the determinants of NAS fees without differentiating between family and non-family firms (Abbott et al., 2003, Abbott et al., 2011, Ye et al., 2011, Zaman et al., 2011). The second stream of research examines the impact of ACE on NAS fees. Extant research provides evidence on the latter association in the US (e.g., Abbott et al., 2003, Abbott et al., 2011), but few studies have investigated this relationship in the UK, where more flexible regulations exist and a different corporate governance system is in place. Moreover, prior research in the UK investigated this association without accounting and controlling for ownership structure (family vs. non-family) and before significant changes were made to the UK Corporate Governance Code through the incorporation of the Smith report in 2003. The third stream investigates family involvement (e.g., Anderson and Reeb, 2003, Wang, 2006, Poutziouris et al., 2015, Kang, 2017). Whereas most of the studies show an association between performance and family involvement in management, directorship, and ownership (active family firms hereafter) (e.g., Anderson and Reeb, 2003, Poutziouris et al., 2015), limited research exists on the influence of family involvement on the purchase of NAS within publicly listed firms.

The results indicate a significant positive association between ineffective ACs and NAS fees. This association is more pronounced for family than for non-family firms, suggesting that in the UK, the former tend to purchase more NAS from their incumbent auditor when they have an ineffective AC. The results also show that these firms pay higher NAS fees when they have ineffective ACs and when family members own shares or hold board positions.

The remainder of the study is organized as follows. Section 2 presents the literature review and hypothesis development, and Section 3 describes the research methodology. Section 4 reports the empirical results, while Section 5 concludes the paper.