This paper provides new evidence on the fee-quality relationship using the propensity to use income-increasing discretionary accruals to meet or beat analysts' forecasts. The evidence in this paper suggests that abnormal audit fees are positively related to audit quality. This result is consistent with concerns raised by regulators that lower audit fees could reflect a lower level of effort provided by the auditor. This is important, given the trend of declining audit fees in recent years. By finding different results using a more focused sample of firms with the incentive and ability to manage earnings, this study highlights the importance of considering the context when performing tests of earnings management. This information is of interest to regulators, such as the SEC.

Eshleman, J. D., & P. Guo. 2014. Abnormal Audit Fees and Audit Quality: The Importance of Considering Managerial Incentives in Tests of Earnings Management. Auditing: A Journal of Practice & Theory 33 (1): 117-138.

In this study, the authors attempt to shed light on the conflicting evidence by performing a study of the relationship between abnormal audit fees and audit quality using a new research design. Specifically, the authors examine whether clients paying abnormal audit fees are more or less likely to use discretionary accruals to meet or beat the consensus analyst forecast.

A growing body of accounting literature examines the relationship between audit fees and audit quality. Researchers are interested in this relationship because, ex ante, it is not clear whether receiving higher fee revenue from a client will improve audit quality or harm it. On the one hand, it could be argued that an auditor who receives abnormally high audit fees from a client will lose their independence and allow the managers of the client firm to engage in questionable accounting practices. However, it is also possible that audit fees are a measure of audit effort, i.e., higher fees indicate that the auditor worked more hours, signaling greater effort. To the extent that audit fees are a measure of audit effort, low audit fees could harm audit quality.

Audit fee and auditor data are obtained from Audit Analytics, financial statement data are obtained from Compustat, and analyst forecast data are obtained from the I/B/E/S database. The authors perform tests on two samples of 4,476 firm-years and 1,670 firm-year observations spanning 2000 to 2011.

The authors find that clients paying higher abnormal audit fees are significantly less likely to use discretionary accruals to meet or beat the consensus analyst forecast. If abnormal audit fees are held at their mean, a one-standard-deviation increase in abnormal audit fees decreases the client's likelihood of using discretionary accruals to meet or beat the consensus forecast by approximately 5 percent. This is consistent with higher audit fees being indicative of greater auditor effort and, ultimately, better audit quality. The authors obtain similar results whether they use the audit fee model of Choi et al. (2010), the one proposed by Blankley et al. (2012), or their own audit fee model.

http://www.auditingresearchsummaries.org

In general, the results of this study provide empirical evidence that capitalized software development costs are incrementally informative about the client’s business risk and the overall audit risk.

• First, although intangible assets play an increasing role in firm valuation, very little is known about whether and how auditors regard these assets in assessing audit risk. Thus it fills a void in the literature on the auditor’s assessment of capitalized software development costs.
• Second, this study examines a setting that mitigates the client’s business risk. It is important in that while economic models predict that audit fees reflect business risk, there is some evidence that audit practice does not support a relation between business risk and audit fees
• Third, this study could be viewed as a ‘‘bridge’’ between the limited literature on investor valuation of capitalized software assets and auditing
• Fourth, this study also shed light on how auditing under certain circumstances could potentially enhance the informativeness of recognized assets that are subject to a high degree of information asymmetry and managerial discretion

For more information on this study, please contact Gopal V. Krishnan.

Krishnan, G. V., and C. Wang. 2014. Are Capitalized Software Development Costs Informative About Audit Risk? Accounting Horizons 28(1): 39-57.

Capitalization of software research and development costs (SDC) is controversial and is the only exception to SFAS No. 2 that calls for immediate expensing of R&D costs. The purpose of this study is to examine whether capitalized SDC are informative about audit risk. Audit risk is the risk that the auditor renders an incorrect opinion on whether the financial statements are in compliance with GAAP.

This sample consists of 564 firm-year observations representing 157 software firms (SIC code 7372) for the period 2004 through 2009. The authors estimate a regression of the log of audit fees on a variety of determinants of audit fees and two measures of lagged capitalized SDC.

• The authors find that on average capitalized SDC has a negative impact on audit fees after controlling for traditional measures of client risk (e.g., negative return on assets, high leverage, going concern opinions, accounting restatements, and internal control problems).
• The results hold for firms where capitalization is inconsequential to beating analysts’ forecasts, firms that are older or have low analysts’ following.
• The study does not find a significant association between audit fees and capitalized software development costs for firms that meet or beat analysts’ forecasts or have high analysts’ following.

The authors claim their results support the notion that capitalized software development costs signal lower business risk—especially for firms with low earnings management risk or high private information (information asymmetry).

http://www.auditingresearchsummaries.org

There are numerous factors that go into a firm’s decision to cross-list on foreign stock exchanges. One factor firms should consider regarding entrance into the U.S. stock exchange is an increase in audit fees. The evidence from this study indicates this increase can be traced back to costs from the legal environment and increased audit effort. 

Bronson, Scott N., A. Ghosh, and C. E. Hogan. 2017. “Audit Fee Differential, Audit Effort, and Litigation risk: An Examination of ADR Firms”. Contemporary Accounting Research 34.1 (2017): 83.

U.S. investors rely on financial statements by foreign firms cross-listed on U.S. stock exchanges. Therefore, these financial statements must comply with accounting standards from the entity’s home country and U.S. standards. Previous studies have identified that audit fees are higher for cross-listed firms and attributed this to added litigation costs. This study examines if there are additional factors causing the audit fees to be higher for cross-listed firms. Specifically, about whether an increase in audit effort is incrementally related to price increases and if audit effort varies based on the stringency of an entity’s home country regulations. The authors presume the additional audit effort will result from the attestation of U.S. GAAP reconciliations and foreign auditor attestation of U.S. audit and independence standards.

The final sample consists of 36,646 observations and only includes entities audited by Big 4 firms. Compustat was used to find U.S.-based publicly traded firms, a list of foreign firms cross-listed in the United States was obtained from Bank of New York Mellon, and foreign non-cross listed publicly traded firms was listed in Worldscope and Compustat Global. The analysis was run using a regression of audit fees.

The authors find the following:

• Cross-listed firms in the United States do in fact pay significantly higher fees relative to other firms.
• This total fee difference includes both incremental costs in legal environment and audit effort. The authors attributed 29% to 48% of this additional incremental costs as the additional audit effort required.
• Cross-listed firms in countries with more stringent audit oversight pay a lower incremental audit fee compared to cross-listed firms in countries with more lax audit oversight.

http://commons.aaahq.org/groups/e5075f0eec/summary

By showing that the social capital where the firm is headquartered affects audit fees, this study makes an important contribution to the auditing literature. It shows that the social environment where the firm is headquartered can affect its relation with auditors and, consequently, the audit fees. This is a new way of looking at the auditor and client’s relation. Although the audit-fee literature is extensive, no studies the authors know of have investigated the possible impact of the social environment on how much the auditors charge. The study suggests that trust is an important component of the auditor-client relation, that local social capital proxies to some extent for auditor trust, and its impact on audit fees is meaningful.

Jha, A., & Chen, Y. 2015. Audit Fees and Social Capital. Accounting Review 90 (2): 611-639.

Social capital is often defined as the mutual trust in society. The authors propose that the social capital in the county where a U.S. firm is headquartered can have an impact on how much the auditors trust the managers of the firm. Auditors arguably have less trust when a firm is headquartered in a county with low social capital. The authors argue that this lack of trust will increase the auditor’s effort and his or her fear of litigation and, therefore, will increase fees. The authors examine the impact of social capital on audit fees.

Audit fees can increase due to more audit work and/or more expected losses. To further investigate which particular element drives up audit fees, the authors examine the impact of social capital on the auditor’s report lag, which is a proxy for the auditor’s effort, and on the firm’s litigation risk, which is a proxy for the auditor’s expected losses.

To test the hypothesis the authors use a multivariate regression. The sample includes 28,634 firm-year observations that represent 5,167 firms and 57 industries, and spans the years 2000 to 2009. The data is collected from Audit Analytics and Compustat. All the firms must be headquartered in the U.S. and belong to a nonfinancial and non-regulated industry.

The audit fees are significantly lower in high social-capital counties. A firm that is headquartered in a county with social capital in the 75th percentile pays about 12 percent less in audit fees compared to a firm headquartered in a county with social capital in the 25th percentile. The auditors take more time to sign off on their report for low social-capital clients. Furthermore, the probability of litigation involving the auditor is also higher in low social-capital counties.

When the auditors are either located within a 100-kilometer (62.13 miles) radius of the client or in the same metropolitan statistical area (MSA) as the client, the effect of social capital on audit fees is tripled compared to when they are further away. Because of the increase in audit complexity due to Sarbanes-Oxley Act (SOX), social capital’s effect is stronger post-2004.These results suggest that auditors take into consideration the social capital of where the firms are headquartered in assessing their audit fees.

The results do not necessarily suggest that auditors are violating professional guidelines that require that they exercise "professional skepticism" in auditing their clients. Rather, the results suggest that the extent of the skepticism can vary based on where their clients are headquartered.

http://www.auditingresearchsummaries.org

This study provides an analysis of initial-year audit fee discounts/premiums in the pre-SOX era (2001) and post-SOX era (2006). 
The findings from this study suggest that concerns over initial-year audit fee discounts are not supported by empirical evidence in the post-SOX era, at least for the Big 4 auditors.  Additionally, this study provides empirical evidence that suggests the Big 4 have become more conservative in their client acceptance and pricing decisions in the post-SOX era.  The authors note that these results
should be of interest to clients, auditors, and regulators, given the concerns expressed by regulators and legislators about the adverse consequences associated with initial-year fee discounts.

Huang, H-W., K. Raghunandan, and D. Rama.  2009.  Audit Fees for Initial Audit Engagements Before and After SOX.  Auditing: A
Journal of Practice and Theory 28 (1): 171-190

The purpose of this study is to examine initial-year audit fee discounts (i.e. “lowballing”) in the post-SOX era.  The study is motivated by concerns (from legislators, regulators, and the media) that auditors “lowball” initial-year audit fees in order to gain the job and that such lowballing can lead to reduced audit quality.  The authors discuss five factors that likely reduced lowballing of audit fees in the post-SOX era: 

• Audit firms are more likely to price auditing as a stand-alone service and less likely to lowball audit fees to gain certain nonaudit service contracts, which are restricted by SOX.
• Audit firms became more conservative in the post-SOX era and the increased conservatism is reflected in first-year fee pricing.
• SOX 301 requires that the audit committee (instead of management) be responsible for compensating auditors. This could reduce client’s audit fee negotiation since the incentives of audit committee members (reputation) can be different from the incentives of management (cost-cutting).
• The failure of Arthur Andersen led to a more concentrated audit market, which in turn changed the dynamics of audit pricing.
• SOX 404 imposed significant additional demands on manpower resources and initial-year engagements, as well as presented a new source of revenues for auditors. 

This study compares audit fees in the pre-SOX and post-SOX periods to determine whether lowballing of audit fees for initial-year audits becomes less likely in the post-SOX era.

The authors use data on U.S. publicly-traded non-financial companies for pre-SOX (2001) and post-SOX (2006) fiscal year-end audits. For each period, they compare audit fees (or change in audit fees) of first-year engagements with those of subsequent-year engagements to determine initial-year fee discounts or premiums. 

The authors examine Big 4 and non-Big 4 clients separately. They exclude auditor resignations and the switches from Big 4 to non-Big 4 (and vice-versa).

• Pricing for initial-year audits for the Big 4 clients is significantly different in the post-SOX era, as compared with the pre-SOX era.
  • Big 4 clients received an initial-year audit fee discount of 24% in 2001.
  • Big 4 clients paid an initial-year audit fee premium of 16% in 2006.

The authors suggest this implies that for the Big 4 auditors, concerns over initial-year discounting of audit fees are not supported by empirical evidence in the post-SOX period. 

• As for non-Big 4 clients, the authors find some evidence (when examining the changes in audit fees from the preceding fiscal year) that non-Big 4 clients received initial-year fee discounts both in 2001 and 2006. The authors suggest this implies that competition is more intense in the non-Big 4 segment of the audit market in the post-SOX period, when compared to the Big 4 segment. 
• Several interesting observations are noted by the authors:
  • Auditor resignations are more likely in 2006 than in 2001.
  • There are very few instances of a Big 4 firm serving as a successor auditor following the resignation of a predecessor auditor.
  • The likelihood of a Big 4 successor following a client-initiated dismissal is lower in 2005-2006 than in 2001.  The authors argue this implies that the Big 4 became more conservative in accepting new clients and/or non-Big 4 became more aggressive following SOX.