Do government-experienced auditors reduce audit quality?

Abstract

Purpose

Recent studies regarding auditor experience generally focus on auditor overall experience in accounting, auditing, finance and related fields (Hardies et al., 2014), auditor sector and domain experience (Bedard and Biggs, 1991; Hammersley, 2006), auditor experience as CPA (Ye et al., 2014; Sonu et al., 2016) or big N experience (Chi and Huang, 2005; Gul et al., 2013; Zimmerman, 2016) or auditors’ international working experience (Chen et al., 2017). But there is little attention paid to where auditors obtained their experience from? And how do auditors with government experience affect audit quality (AQ)? This paper aims to present the effect of auditors with government experience on AQ.

Design/methodology/approach

The authors used Turkish publicly traded firms in Borsa Istanbul between the year 2008 and 2015 to test the hypothesis. The sample comprises 1,067 observations and eight years. Two main proxies of government experience are used in this paper. The first proxy is auditor’s government experience in the past. The second proxy is the continuous variable which is “the logarithmic value of the number of years of government experience”. Further, auditor overall experience in auditing, accounting, finance and other related fields are also used as a control variable. Audit reporting aggressiveness, audit reporting lag and discretionary accruals are used as proxies of AQ. Besides this, the authors adopted the model to estimate the probability of selecting a government-experienced auditor, and they presented the regression results with the addition of inverse Mills ratio.

Findings

The main findings are consistent with conjecture. Government-experienced auditors do not enhance AQ. They are aggressive, and they complete audit work slowly and they cannot detect discretionary accruals effectively. Spending more time in a government agency makes them more aggressive and slow, and they do not detect earnings management practices. The Heckman estimation results regarding the variable of interest are also consistent with the main estimation results. In addition, the authors found in predicting government-experienced auditor choice that family firms, domestic firms and firms that reported losses (larger firms, older firms) are more (less) likely to choose government-experienced auditors.

Research limitations/implications

This study has some limitations. The authors used a small sample to test the impact of government-experienced auditors on AQ because of data access problems. Much data used in this study were collected manually. Earnings quality was calculated using only discretionary accruals. Real activities manipulation was not used as the proxy of AQ in this paper. The findings from emerging markets might not generalize to the developed countries because the Turkish audit market is developing compared to Continental Europe or USA.

Practical implications

The findings are considered for independent audit firms. Audit firms may employ new graduates and train them to offer more qualified audit work for their clients. The results do not mean that government-experienced auditors should not work in an audit firm, or that they should not establish an audit firm. It is clear that government-experienced auditors provide low AQ in terms of audit reporting aggressiveness, audit report lag and discretionary accruals. But as they operate more in the independent audit sector, they will become successful and provide qualified audit work. One other thing we can say is that it is perhaps better for government-experienced auditors to work in the tax department of independent audit firms.

Originality/value

This paper tries to fill the gap in the literature regarding the effect of auditor experience on AQ and concentrates on a different type of experience: Auditors with government experience.

Citation

Ocak, M. and Can, G. (2019), "Do government-experienced auditors reduce audit quality?", Managerial Auditing Journal, Vol. 34 No. 6, pp. 722-748. https://doi.org/10.1108/MAJ-12-2017-1756