Audit Partner Characteristics, Audit Quality and Audit Pricing in - - PowerPoint PPT Presentation

Audit Partner Characteristics, Audit Quality and Audit Pricing in the U.S.

Ally Zimmerman, Northern Illinois University, azimmerman5@niu.edu Al Nagy, John Carroll University, alnagy@jcu.edu 2016 Deloitte/KU Audit Symposium, May 20

8/16/2016 1

Introduction

• RQ: Do audit fees and audit quality vary among U.S. audit partners with partner

experience and gender?

• Studies from other countries have found audit partner effects, but evidence is mixed
• U.S. audit market different from other countries
• Litigious environment, regulatory scrutiny/inspections
• No Disclosure of audit partner identity to investors
• No U.S. evidence yet - Partner identity disclosure will be required in the U.S. starting

in 2017 on PCAOB Form AP

• Contribution
• Understanding the impact of individual auditors on audit outcomes over and above

their audit firms/offices

• What can data from the period 2004-2014 prior to disclosure of audit partner

identity tell us about the impact of audit partner characteristics on audit pricing, quality?

• Evidence pre- and post-disclosure can be compared in the future to isolate the

impact of audit partner identity disclosure itself

8/16/2016 2

Prior Research Outside of U.S.

• In Australia, Taylor (2011) finds that premium partners audit fewer clients than the

discount partners, but on the contrary have shorter partner tenures.

• In Sweden, partner industry specialization is positively associated with audit fees

(Zerni 2012).

• Ittonen and Peni (2012) find that firms with female audit engagement partners have

significantly higher audit fees.

• Gul et al. (2013) identify significance of audit partner effects on audit outcomes in

China.

• In Australia, Goodwin and Wu (2014) find that audit fee premiums are associated with

partner-level industry expertise.

• Ye, Cheng, and Gao (2014) document that individual auditors with more auditing

experience are less likely to be associated with audit failure.

• Hardies et al. (2015) also find a female audit fee premium among Belgian public

companies.

• Cahan and Sun (2015) find that experience is positively associated with audit fees in

China.

• Wang et al. (2015) find that, in some models of restatements, engagement and review

partner experience is positively related to restatements (audit failures).

8/16/2016 3

Does Partner Experience Matter in U.S.?

• Audit partner identity matters to management and

audit committees in the auditor selection process

• Partners/their resumes are heavily vetted in the

auditor selection process.

• Technical expertise as well as political capital within an

audit firm arguably takes years to develop.

• Technical/industry expertise is associated with higher

audit quality

• If partner political capital and expertise are valuable to

clients, they would be willing to pay more for it through higher audit fees. H1a/b: Audit partner experience is positively associated with audit fees and audit quality.

8/16/2016 4

Does Partner Gender Matter?

• Females executives tend to be more diligent, more

conservative, less overconfident and less tolerant of risk than males (e.g., Palvia et al. 2015; Huang and Kisgen 2013; Peni and Vahamaa 2010; Eagly and Carli 2003) -> Higher audit quality

• Female tolerance for less risk may affect the pricing

decisions by increasing auditor effort.

• If females tolerate less risk, they may also charge a

premium for taking on this risk. H2a/b: Female audit partners have higher audit fees and higher audit quality than male audit partners.

8/16/2016 5

Data and Sample

• We identified through Audit Analytics SEC Comment Letter Database a

total of 6,562 SEC comment letter correspondences from 2004 to Oct 2014 that copied 1,750 unique Big 4 audit partners

• Correspondences could copy more than one audit partner; if more than
• ne partner copied, assumed the first person copied was the lead partner
• We assumed that the copied audit partner was an engagement partner on

the audit of the firm in the year of the letter (alternatively the prior year based on CompuStat year)

• Partner background information hand-collected from LinkedIn
• Excluded financial firms, foreign-based, those missing model variable

information and who partner information could not be located

• Final sample for analysis – 769 firm-year observations encompassing 479

unique firms and 471 unique partners – Some firms and partners but not all have multiple appearances

8/16/2016 6

Example SEC Correspondence Copying the Auditor

8/16/2016 7

Audit Pricing Model

AUFEE = β0 + β1EXPER + β2GENDR + β3ASSETS + β4ROA + β5LEV + β6LOSS + β7NSUBS + β8REC + β9INV + β10GC + β11FORSA + β12DECYE + β13S404 + β14SPEC + + β15TEN + β16-23IND + β24-33YEAR,

EXPER = number of years the partner is with the audit firm or the number of years the partner has been in practice (number of years elapsed since the bachelor’s/master’s degree). GENDR = 1 if partner is female, 0 if male. ASSETS = log of total assets. ROA = ratio of net income to total assets. LEV = ratio of long-term debt to total assets. LOSS = 1 if company had a net loss in last 3 years, else 0. NSUBS = log of the number of consolidated subsidiaries. REC = ratio of receivables to total assets. INV = ratio of inventory to total assets. GC = 1 if going concern modification in the audit report, else 0. FORSA = ratio of foreign sales to total sales. DECYE = 1 if December fiscal year-end, else 0. S404 = 1 if integrated SOX 404(b) and external audit is conducted, else 0. SPEC = 1 if auditor is national specialist per SIC code (2-digit), else 0. TEN = 1 if auditor tenure is less than 3yrs

8/16/2016 8

Variablea MEAN MEDIAN STD DEVIATION

AUFEE 0.473 0.352 1.105 EXPER 19.17 19.00 8.195 GENDR 0.113 0.000 0.317 ASSETS 6.939 6.844 1.984 ROA

• 0.864

3.450 14.616 LEV 0.203 0.146 0.244 LOSS 0.468 0.000 0.499 NSUBS 0.548 0.000 0.689 REC 0.126 0.104 0.103 INV 0.092 0.047 0.118 GC 0.417 0.000 0.493 FORSA 0.221 0.020 0.286 DECYE 0.620 1.000 0.486 S404 0.805 1.000 0.397 SPEC 0.247 0.000 0.432 TEN 0.108 0.000 0.310

Descriptive Statistics Audit Fee Analysis

8/16/2016 9

Audit Pricing OLS Regression Results

AUFEE = β0 + β1EXPER + β2GENDR + β3ASSETS + β4ROA + β5LEV + β6LOSS + β7NSUBS + β8REC + β9INV + β10GC + β11FORSA + β12DECYE + β13S404 + β14SPEC + β15TEN + β16-23IND + β24-33YEAR

Variable Prediction Coefficients t- statistic EXPER + 0.006 2.68*** GENDR + 0.119 1.92** ASSETS + 0.485 34.17***

Number of Observations 769 Adjusted R2 77.28% Model F-Value 82.66***

8/16/2016 10

Audit Quality Analysis

GC = β0 + β1EXPER + β2GENDR + β3LSPEC + β4OFFSIZE + β5PBANK + β6ASSETS + β7AGE + β8LEV + β9CLEV + β10LLOSS + β11INVEST + β12FEERATIO + β13CFFO + β14TENURE + β15-20IND + β21-29YEAR

LSPEC = 1 if auditor has highest local market share defined as two-digit SIC code within the metropolitan statistical area (MSA), else 0; OFFSIZE = natural log of the total local office audit fees other than the observation; PBANK = probability of bankruptcy measured by adjusted Zmijewski score; ASSETS = natural log of total assets; AGE = natural log of the number of years included in Research Insight; LEV = ratio of total liabilities to total assets; CLEV = change in LEV during the year; LLOSS = 1 if company reported a loss for the previous year, else 0; INVEST = short- and long-term investment securities (measured as current assets less receivables and inventory) divided by total assets; FEERATIO= ratio of non-audit fees to total fees paid to the incumbent auditor; CFFO = cash flow from operations divided by total assets; and TENURE = natural log of years the audit firm on the engagement.

8/16/2016 11

Variablea MEAN MEDIAN STD DEVIATION

GC 0.401 0.000 0.492 EXPER 18.10 18.00 7.872 GENDR 0.105 0.000 0.307 LSPEC 0.424 0.000 0.496 OFFSIZE 17.94 17.98 0.916 PBANK

• 3.198
• 3.648

2.247 ASSETS 5.424 5.203 1.637 AGE 2.141 2.303 0.831 LEV 0.538 0.474 0.393 CLEV

• 0.029

0.016 0.599 LLOSS 0.750 1.000 0.434 INVEST 0.453 0.404 0.289 FEERATIO 0.122 0.086 0.118 CFFO

• 0.090
• 0.004

0.264 TENURE 1.703 1.946 0.779

Descriptive Statistics (N=172) Going Concern Analysis Sample

8/16/2016 12

Going Concern Analysis Results

Number of Observations 172 Pseudo R2 59.43% Model Chi-square 135.57***

GC = β0 + β1EXPER + β2GENDR + β3LSPEC + β4OFFSIZE + β5PBANK + β6ASSETS + β7AGE + β8LEV + β9CLEV + β10LLOSS + β11INVEST + β12FEERATIO + β13CFFO + β14TENURE + β15-20IND + β21-29YEAR

Estimated Wald Variable Prediction Coefficients Chi-square EXPER + 0.103 5.52*** GENDR +

• 0.428

0.19 LSPEC + 1.317 3.67**

8/16/2016 13

Results Summary

• Strong Results for Audit Pricing

– More experienced audit partners associated with higher audit fees – Female audit fee premium

• Mixed Results for Audit Quality

– More experienced partners are more independent

• > more likely to issue a going concern modified
• pinion for distressed companies

– No difference in audit quality/ independence with respect to partner gender

8/16/2016 14

Conclusion, Contribution, Implications

• We provide initial evidence that audit pricing and audit quality do vary

with audit partner characteristics in the U.S., indicating that partner identity could be valuable information to investors and supporting the PCAOB’s initiative to have firms disclose audit partner identity.

• Following partner identity disclosure, partners will likely develop

reputations separate from their audit firms – Whose reputation will matter more – partners or firms?

• Our results for audit pricing and audit quality are based on an

environment pre-public disclosure of audit partner identity. Results using data post-public disclosure of audit partner identity can be compared to results of the paper to isolate the effect of disclosure on the audit market in the U.S.

• Future research to understand the underlying drivers of the experience

and gender effects that we find.

8/16/2016 15

Future Analyses

• Checking for nonlinear relation (initial partner

period vs. nearing retirement)

• Isolate partner experience from client

experience with the partner’s audit firm

• Additional audit quality analyses

8/16/2016 16

THANK YOU to Deloitte and KU Send comments to: Ally Zimmerman azimmerman5@niu.edu

8/16/2016 17