Do charitable auditors deliver better audit quality? Evidence from Chinese CPAs

ABSTRACT

This paper investigates the relationship between auditors’ participation in charity and audit outcomes. Using charitable activity records of Chinese CPAs, we find that auditors who engage in charitable activities are associated with significantly better audit quality. Specifically, we examine engagement auditors and review auditors separately and find that engagement auditors with records of charitable activities demonstrate lower discretionary accruals and are more conservative in issuing modified audit opinions. However, firms with review auditors who engage in charitable activities are associated with fewer restatements. Taken together, our results suggest that an individual auditor’s charity activity is associated with audit quality.

KEYWORDS:

• Intrinsic motivation
• codes of ethics/professional conduct
• individual auditors
• audit quality

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. However, we believe this is less likely the case in China for the following reasons. First, even in face of internal reviews and controls, prior studies still find significant evidence that audit quality vary between audit firms of different sizes, suggesting such mechanisms does not deter audit quality variation (DeFond, Erkens and Zhang 2017)2017, at least not among firms of different sizes. This actually is part of the reasons why external monitoring such as PCAOB oversight are required even after the regulatory reform of SOX in the early 2000 amid the series high-profile accounting scandals involving corporations like Enron, WorldCom, etc. As for the external monitoring, China’s monitoring and legal environment has always been considered weak compared with the that of the western countries

2. Lennox, Wang, and Wu (2020)find that the asscoiation between equity ownership and audit adjustments, which reverses the earnings management, are significantly positive for engagement auditors and negative or insignificant for the review auditors. This is consistent with the moniotring role of review auditors and with the business-seeking characteristics of engagement auditors. It is possible that our results on the association between auditors’ charitable activities and audit quality varies between the two groups of auditors.

3. He, Pan, and Tian (2017) find that audit firms after completing the transformation demonstrate better audit quality in terms of conservatism in issuing modified opinion, earnings management and premium in audit fees.

4. Zeng et al. (2021) find that audit quality improves after the disclosure of key audit matters.

5. Huang (2013) find that only a total of sixty-five securities civil cases are brought by investors due to misrepresentation during the period from 2002 to 2011 after the PSL rules become effective, in comparison with that over the same period there are 253 cases of criminal judgment or administrative sanction.

6. For example, Chen et al. (2016) find that the monetary fines was just equal to the audit fees charged by the audit firms when the audit failure happened, and even in the most severe circumstance of revoking a license the audit partners of an audit firm can minimize their damages by leaving the current audit firms and taking their clients to a new audit firm.

7. The Chinese audit market is much more competitive and dispersed than the U.S. and other developed countries, where the Big 4 auditors occupy the majority of the audit market (Chen, Sun, and Wu 2010).During our sample period of 2003–2015, the proportion of publicly-listed firms per year audited by the Big 4 audit firms is 4.06%(4.09%) for the test of audit quality by engagement auditors (review auditors)

8. The examples of policies an audit firm can adopt include screening new clients and existing clients, employing training materials, industry-specific databases, internal benchmarks for best practices, automatic IT systems, and internal consultative practices.

9. Kohlberg (1969, pp. 382–391) is a pilot study that proposes a theoretical framework of moral reasoning in which the author identifies three levels of moral development, each consisting two stages. The basic level of moral reasoning (Level 1), the ‘pre-conventional’ level, is characterized as being guided by fear of punishment (Stage 1) or by self-interest, i.e. satisfying one’s own needs (Stage 2); the intermediate level of moral reasoning (Level 2, the ‘conventional’ level) is characterized as being guided either by the wish to please or help others (Stage 3) or by the desire to fulfil societal, legal or religious norms, i.e. ruled-based reasoning (Stage 4); the highest level of moral reasoning (Level 3, the ``post-conventional’’ level) is characterized as being guided by either general, individual rights and socially accepted standards, i.e. a social contract (Stage 5) or from universal ethical principles as defined by the decision-maker’s conscience (Stage 6).

10. See discussion of studies reported in Glazer and Konrad (1996)1996. Just as the words of Adam Smith (17591759) point, they make moral decisions by assessing their own conduct through the eyes of an ‘impartial spectator,’ an ‘ideal mate within the breast’: ‘We endeavor to examine our own conduct as we imagine any other fair and impartial spectator would examine it. If, upon placing ourselves in his situation, we thoroughly enter into all the passions and motives which influenced it, we approve of it, by sympathy with the approbation of this supposed equitable judge. If otherwise, we enter into his disapprobation, and condemn it.’

11. Level 1 refers to the ‘preconventional’ level where moral reasoning is guided either by fear of punishment (Stage 1) or by self-interest, i.e. satisfying one’s own needs (Stage 2). Level 2 is the ‘conventional’ level where moral reasoning is guided either by the wish to please or help others (Stage 3) or by the desire to fulfil societal, legal or religious norms, i.e. ruled-based reasoning (Stage 4). Level 3 is the ``postconventional’’ level, where moral reasoning derives from either general, individual rights and socially accepted standards, i.e. a social contract (Stage 5) or from universal ethical principles as defined by the decision-maker’s conscience (Stage 6).

12. In the DIT subjects are given some moral dilemmas questions and asked to rate and rank the relative moral importance of each of several standard items for solving the dilemma on a 5-point scale.

13. Following prior research by Gul, Ma, and Lai (2012), we distinguish review auditors from engagement auditors based on the size of client portfolios of each signatory auditor. Usually, review auditors are associated larger client portfolios in terms of client firms’ assets. In CSMAR, there are firm-years with two or three auditors with the same client portfolio sizes. We exclude these observations from our sample because we are not able to separate them into review vs. engagement auditors based on client portfolio sizes.

14. We use variables in the CSMAR to calculate the five audit-quality measures except for restatements (for the discussion of audit- quality measures, please see section 4.2.1). For restatements, we manually collect from the corporate disclosures regarding restatements in the China Information website (http://www.cninfo.com.cn). In addition, in annual financial reports where firms disclose financial restatement information in the section called “The causes for and effects of significant accounting errors, a publicly listed firm also discloses financial restatement information in the section called ‘The causes for and effects of significant accounting errors.’ Therefore, we download annual financial statements and corporate disclosures regarding restatements from China and manually collect the causes for restatements, the restated RMB amounts, and the periods for restatements. The sample period for restatements we collected is from 2003 to 2013.

15. .We conducted robustness check using the frequency of the charitable activities. Untabulated results show that our results are qualitatively similar in the sense that for the review auditors, the frequency of charitable activities is associated with restatement; and for the engagement auditors, the frequency of charitable activities is associated with the frequency of meet or beat analyst forecasts. These results suggest that the more likely the auditors to have charitable engagement, the higher audit quality

16. Ho et al. (2007) recommend that before parametric estimation, matching could be used as a preprocessing technique to eliminate model misspecification. Both the linear and nonlinear relation between the treatment variable and the control variables in the population is eliminated in the matched samples, and this mitigates the model misspecification problem. For example, after matching clients audited by charitable auditors and by non-charitable auditors on client size, the linear and non-linear correlation (e.g. polynomial relationship) between the Charity and Size variable disappears, and the nonlinear effect of size is less likely to be captured by the Charity variable. A key advantage of matching is that it does not require identifying exogenous variables or exclusion restrictions (e.g. variables uncorrelated with the main outcome variable) in predicting the treatment choice.

17. We also obtain the information of charitable activities conducted by audit firms from the enquiry system compiled by the Chinese Institute of Certified Public Accountants (CICPA). Audit firms in China are also required by CICPA to disclose their charitable activities through the enquiry system. We find that 30% of audit firm years have records of charitable activities. We acknowledge that the engagement (review) auditors’ participation in charitable activities organized by their audit firms may not be voluntary and can’t reflect their intrinsic motivation. But those charitable activities organized by the audit firms only increase the noise of our measurement and are supposed to go against our findings. Moreover, we can identify whether a charitable activity done by an individual auditor is organized by his/her audit firm because the information about the charitable activity from the enquiry system in CICPA also indicates who organizes the charitable activity. So we exclude those charitable activities organized by the audit firm and find the results are quantitatively the same