business strategy, financial reporting irregularities, and audit...

Business Strategy, Financial Reporting Irregularities,

and Audit Effort*

KATHLEEN A. BENTLEY, University of New South Wales

THOMAS C. OMER, Texas A&M University

NATHAN Y. SHARP, Texas A&M University

1. Introduction

This study uses Miles and Snow’s 1978 and 2003 organizational strategy typology toexamine whether companies that follow different business strategies exhibit differences inthe occurrence of financial reporting irregularities and whether firms’ business strategiesare a factor in determining the audit effort necessary to attest to the firms’ financial state-ments. By exploring the extent to which firms following particular business strategies aremore likely to experience financial reporting irregularities, we provide evidence thatincreases our understanding of the underlying determinants of financial reporting quality.By examining the relation between business strategy and audit effort, we also provide evi-dence on the extent to which audit firms appear to incorporate business strategy in devel-oping their audit planning. We develop a measure of business strategy based on Miles andSnow 1978 and 2003 and test the association between this business strategy measure andseveral proxies for financial reporting irregularities as well as its contribution to an auditfee model generated from prior literature.

Investigating the extent to which firms that follow particular business strategies are morelikely to experience financial reporting irregularities is consistent with a call for research inZahra, Priem, and Rasheed 2005 (813) who suggest that “accounting research, by and large,has focused on identifying potential indicators or ‘red flags’ rather than establishing directcauses or antecedents [of misreporting]”. We discuss the business risks of certain businessstrategies in the context of SAS No. 99 (American Institute of Certified Public Accountants[AICPA] 2002) in order to describe how business strategy may be an underlying determinantof both the occurrence of financial reporting irregularities and additional effort by auditorsin the attestation of financial statements.

Considering business strategy and its association with audit effort is also timely in thecontext of AU 311 (AICPA 2006), which refers to the need for auditors to understand theclient’s industry and business in planning an audit, and ISA 315 (International Federationof Accountants [IFAC] 2009), which emphasizes a business risk assessment in the auditplanning phase and directs auditors to obtain an understanding of “the entity’s objectives

* Accepted by Steven Salterio. This study was the recipient of a 2010 PwC INQuires Grant. We thank Steven

Salterio, two anonymous reviewers, Urton Anderson, Max Baker, Leonard Bierman, Elizabeth Carson,

Demetris Christodoulou, Rajib Doogar, Neil Fargher, Peter Gillett, Shirley Gregor, Carl Hollingsworth,

Andrew Jackson, Kerry Jacobs, Karim Jamal, Shane Johnson, Christine Jubb, Bill Kinney, Robert Knechel,

Habib Mahama, Mary Lea McAnally, Stuart McLeay, Gary Monroe, Ed O’Donnell, Mark Peecher, Joshua

Ronen, Greg Shailer, Marjorie Shelley, Ira Solomon, Ken Trotman, Senyo Tse, Dechun Wang, Mike Wil-

kins, Mark Wilson, and seminar participants at the 2011 AAA Auditing Section Midyear Conference, Uni-

versity of Illinois 19th Audit Symposium, Australian National University, Texas A&M University,

University of New South Wales, and University of Sydney for helpful comments. We are grateful to

PricewaterhouseCoopers, Texas A&M University, University of New South Wales, and the Ernst & Young

Professorship for financial support.

Contemporary Accounting Research Vol. 30 No. 2 (Summer 2013) pp. 780–817 © CAAA

doi:10.1111/j.1911-3846.2012.01174.x

and strategies, and those related business risks that may result in risks of materialmisstatement” (ISA 315.11(d)). In addition, AS No. 5 has encouraged auditors to take atop-down, risk-based approach to internal control audits, which are scaled in relation tothe client’s size and complexity (Public Company Accounting Oversight Board [PCAOB]2007). AS No. 5 emphasizes a “risk-based approach” because “the size and complexity ofthe company, its business processes, and business units, may affect the way in which thecompany achieves many of its control objectives” (PCAOB 2007, AS 5.13). Thus, thisstudy provides evidence on whether business strategy, a contributing factor in the size andcomplexity of the company and its business processes, is associated with varying degreesof audit effort as reflected in an audit fee model.

Miles and Snow (1978, 2003) detail three viable business strategies that may exist simul-taneously within industries—Prospectors, Defenders, and Analyzers—where the key dimen-sion of the strategy typology is the organization’s rate of change regarding its products andmarkets (Hambrick 1983).1 These strategies are positioned along a continuum, with pros-pectors at one end and defenders at the other. Prospectors rapidly change their product-market mix to be innovative market leaders in numerous domains, while defenders maintaina narrow and stable product focus to compete on the basis of price, service, or quality(Miles and Snow 1978, 2003). Firms that constitute the middle of the continuum are analyz-ers, which have attributes of both prospectors and defenders (Miles and Snow 1978, 2003).Consistent with prior research in both management and accounting (e.g., Hambrick 1981,1983; Simons 1987; Ittner, Larcker, and Rajan 1997), we focus on the two distinct strategiesat the ends of the continuum, prospectors and defenders. Appendix 1 provides a moredetailed description of both the prospector and defender strategies.

Our main findings suggest that, ceteris paribus, the odds of experiencing an Accountingand Auditing Enforcement Release (AAER), lawsuit, or restatement are 2.32, 1.10, and0.47 times higher, respectively, for firms with a STRATEGY score at the cutoff for prospec-tors relative to firms with a STRATEGY score at the cutoff for defenders. In addition, we findthat, ceteris paribus, firms at the cutoff for prospectors will pay audit fees approximately 22percent higher than firms at the cutoff for defenders, suggesting that auditors incorporate acompany’s business strategy in their audit plan and that audit effort varies across businessstrategies. We examine several possible explanations for why prospectors experience a greaterlikelihood of irregularities despite the apparent increase in auditor effort. Our findings sug-gest that the higher audit fees (and hence higher audit effort) for prospectors are insufficientto address the riskiness of these clients. Additional sensitivity tests suggest that our strategymeasure captures client business risk and that the observed increase in audit effort for pros-pector firms is incremental to effects related to financial reporting risk or other risk-basedmeasures used in the audit fee literature. Specifically, business strategy is positive and signifi-cant in the audit fee model in the presence of financial reporting risk factors, and it is distinctfrom measures commonly used to explain the distribution of audit fees. Together, this evi-dence suggests our measure of business strategy is a separate construct providing incrementalinformation content beyond traditional measures of client size, risk, and complexity.

Our study links three research literatures: organizational theory from the managementliterature, and financial reporting quality and audit effort from the accounting literature. Ourcontributions are fourfold. First, we develop a comprehensive measure for organizationalbusiness strategy that is generalizable across industries and easily replicable. Second, whileprior accounting research has examined business strategy as a determinant of compensation(Ittner et al. 1997), accounting control systems (Simons 1987), budgetary usage (Collins,Holzmann, and Mendoza 1997) and in contemporaneous research as a determinant of tax

1. Miles and Snow (1978, 2003) indicate that, although a fourth business strategy exists (Reactors), this strat-

egy is not viable in the long term and is often difficult to identify. Thus, we focus on the viable strategies.

Business Strategy, Financial Reporting Irregularities, and Audit Effort 781

CAR Vol. 30 No. 2 (Summer 2013)

aggressiveness (Higgins, Omer, and Phillips 2011), we contribute to this line of research byproviding evidence that business strategy has an even broader application to financial report-ing and auditing than previously considered. Third, by relying on organizational theory tocategorize companies and their associated business risk, we develop a better understandingof the factors that are ex ante determinants of financial reporting irregularities. Finally, ourresults provide evidence that auditors, consistent with authoritative guidance, appear to takea broader, more comprehensive approach in assessing their audit clients’ business risk thanwhat has been represented in the literature using traditional proxies such as client size andcomplexity. However, improving audits to reduce financial reporting irregularities amongprospector clients appears to be an important area for audit practice and future research.

The remainder of our paper is organized as follows. Section 2 discusses Miles andSnow’s 1978 and 2003 business strategy types, and section 3 develops our hypotheses. Sec-tion 4 describes our measures and models. Section 5 describes our data. Section 6 describesour empirical results, while section 7 presents additional analyses. Section 8 concludes.

2. Business strategy

The management literature provides several typologies of business strategy that describehow companies compete in their respective market environments. Some of the more well-known typologies, in addition to Miles and Snow 1978 and 2003, include: Porter 1980,who describes business strategies in terms of cost leadership and product differentiation;March 1991, who describes business strategies in terms of exploration and exploitation;and Treacy and Wiersema 1995, who describe business strategies in terms of operationalexcellence, product leadership, and customer intimacy. While the labels for business strate-gies differ across the various typologies, a common feature of all the proposed strategyclassifications is that they most clearly identify companies that operate at one end or theother of a strategy continuum (Dent 1990; Langfield-Smith 1997; Seifzadeh 2011). Thus,while each typology makes some attempt to classify companies that operate under a mixedstrategy, these mixed strategies generally exhibit, to varying degrees, the characteristics ofcompanies at both ends of the strategy continuum.

Although Miles and Snow’s theory proposes four business strategies (three of whichare viable), consistent with prior research in management and accounting we focus ourdiscussion on the two distinct strategies that comprise the endpoints of their strategy con-tinuum. Miles and Snow refer to these companies as prospectors and defenders.2 The char-acteristics of prospectors and defenders are common to the business strategies suggestedby Porter, Treacy and Wiersema, and March (Dent 1990; Langfield-Smith 1997; Seifzadeh2011). Specifically, Miles and Snow’s prospector strategy aligns with Porter’s Product Dif-ferentiation, March’s Exploration, and Treacy and Wiersema’s Product Leadership; andMiles and Snow’s defender strategy aligns with Porter’s Cost Leadership, March’s Exploi-tation, and Treacy and Wiersema’s Operational Excellence.

We select the Miles and Snow classification for two reasons. First, given the common-alities across the various business strategy typologies noted above, inferences based onMiles and Snow are likely to align with inferences based on the other classifications. Secondand more importantly, while the Miles and Snow typology can be operationalized usingarchival data (e.g., Ittner et al. 1997), the other typologies require personal interviews andsurveys of corporate officers. Thus, our methodology produces a replicable measure ofbusiness strategy that allows us to generalize our results to a broad cross-section of compa-nies and industries. We note that much of the detailed discussion of prospectors anddefenders below is transferable to the business strategies proposed by alternative theories.

2. Companies following the third viable strategy, analyzers, have attributes of both prospectors and defenders

and thus lie between prospectors and defenders on the strategy continuum.

782 Contemporary Accounting Research


Miles and Snow (1978, 2003) define prospectors as innovative companies seeking toidentify and exploit new products and market opportunities; thus, prospectors have bud-gets that are oriented toward research and development (R&D) and marketing. Thisfocus on innovation requires prospectors to develop multiple technologies for a diverseproduct mix. Technological flexibility allows prospectors to respond rapidly to change,but it also comes at the cost of rarely achieving maximum efficiency in their productionand distribution. Because prospectors grow through product and market development,Miles and Snow (1978, 2003) suggest that growth may occur in spurts. In order to facili-tate and coordinate prospectors’ diverse and numerous operations, control is decentral-ized; however, Miles and Snow (1978, 2003) highlight the transitory nature of the“dominant coalition”, which lends itself to organizational instability. Prospectors avoidlengthy commitments to a single technological process by maintaining a low degree ofmechanization or routinization and by leveraging the knowledge and skills of theiremployees.

Miles and Snow (1978, 2003) define defenders as companies focused on efficiency inthe production and distribution of goods and services. Due to their narrow market focus,defenders develop closely related products and services rather than pursuing new productand market opportunities, which limits their product development efforts. Defenders “pro-tect” the finance and production functions compared to prospectors’ efforts to “protect”the marketing and research and development functions. Unlike prospectors, defendersgrow cautiously and incrementally through market penetration and hence demonstratelow, steady growth. Defenders are characterized as maintaining strict centralized organiza-tional control to ensure efficiency; Miles and Snow (1978, 2003) indicate that these compa-nies tend to have lengthy employee tenure and promote from within. Achievingproduction and distribution efficiency requires defenders to invest heavily in technologicalefficiency, which includes focusing on “single core” cost-efficient technology and continualimprovement leading to routinization and mechanization.

In the next section we develop our hypotheses about how these business strategiesrelate to both financial reporting irregularities and audit effort.

3. Hypothesis development

Business strategy, business risk, and financial reporting irregularities

Consistent with prior research (e.g., Johnstone 2000; Johnstone and Bedard 2003; Stanley2011), we define client business risk as “the risk that the audit client’s economic conditionwill deteriorate in the future” (Stanley 2011: 157). The auditor’s assessment of client busi-ness risk critically impacts audit planning and effort due to its link to both audit risk (i.e.,the risk that an unqualified audit opinion is issued for a set of materially misstated finan-cial statements) and auditor business risk (i.e., the auditor’s risk of loss in connection withthe audited financial statements—e.g., due to litigation, loss of reputation, etc.).3 SAS No.99 considers fraud in the context of a financial statement audit, and its framework isapplicable to the broader set of financial reporting irregularities we examine in this study.Specifically, the framework suggests that three factors are usually present when financialreporting irregularities occur: incentive, opportunity, and rationalization (AICPA 2002).Prior literature (see Hogan, Rezaee, Riley, and Velury 2008) has provided substantialempirical evidence on the first two risk factors (incentive and opportunity); hence, weorganize our discussion around these two factors.4

3. Refer to Johnstone and Bedard 2003 for a thorough discussion.

4. Although the SAS No. 99 framework suggests three factors are present when financial reporting irregulari-

ties occur, we do not explore the rationalization factor because the literature does not provide a reasonable

archival proxy for this factor. Using survey data, Bentley (2012) develops a measure that proxies for ratio-

nalization of financial statement irregularities in the context of business strategies.



Incentives

Prior research suggests the likelihood that firms engage in financial reporting irregularitiesincreases with such incentive factors as “rapid growth, compensation incentives, stockoptions, the need for financing, and poor performance” (Hogan et al. 2008: 246). Regard-ing growth patterns, SAS No. 99 indicates that “rapid growth ...especially compared tothat of other companies in the same industry” creates an incentive for companies to mis-state their financial results (AU 316.85 [AICPA 2002]). Prospectors display rapid and spo-radic growth patterns, which stem from being innovative and “first to market” in a broadarray of product-market domains (Miles and Snow 1978, 2003). Conversely, defenders dis-play cautious and incremental growth patterns where they grow primarily through marketpenetration of their narrowly focused product line (Miles and Snow 1978, 2003). Thus inlight of SAS No. 99 and prior research (Hogan et al. 2008), prospectors have a greater riskof financial misreporting due to their rapid and sporadic growth tendencies compared todefenders.

SAS No. 99 suggests that firms that base a significant portion of managements’remuneration (e.g., bonuses or stock options) on achieving aggressive performancetargets have greater incentives to misreport (AICPA 2002). In particular, prior researchhas linked stock-option–based compensation to greater occurrences of financial misrep-orting (e.g., Beneish 1999; Burns and Kedia 2006; Efendi, Srivastava, and Swanson2007). Miles and Snow (1978, 2003) suggest that business strategy impacts rewardstructures, and prior research provides support for reward structure differences betweenprospectors and defenders (Ittner et al. 1997; Rajagopalan 1997; Singh and Agrawal2002).

Prospectors’ focus on innovation produces greater outcome uncertainty and thusrequires compensation contracts that not only encourage risk-taking but also take alonger-term perspective, allowing innovative ideas to yield profitable results (Rajagopalan1997). Rajagopalan (1997) and Singh and Agrawal (2002) find that prospectors emphasizelong-term, stock-based compensation incentives (e.g., stock options) to encourage risk-tak-ing behavior in managers. In addition, Singh and Agrawal (2002) indicate that, relative todefenders, prospectors’ compensation packages are less likely to emphasize a fixed paycomponent (e.g., salary) and note that in some prospector firms the entire compensationpackage relies solely on the firm’s market performance (i.e., there was no annual compen-sation component). Similarly, Simons (1987) reports that prospectors, relative to defend-ers, have a larger percentage of managers’ total remuneration linked to aggressiveperformance bonuses. Altogether, prospectors’ compensation arrangements appear to pro-vide stronger incentives for management to misreport financial results because prospectorsbase a larger proportion of management’s remuneration on stock-based compensation,which prior research links to greater occurrences of financial misreporting (e.g., Beneish1999; Burns and Kedia 2006; Efendi et al. 2007).

On the other hand, defenders’ focus on efficiency (e.g., to compete on the basis ofproduct cost) produces less outcome uncertainty and allows compensation contracts totake a shorter-term perspective (Rajagopalan 1997). Rajagopalan (1997) and Singh andAgrawal (2002) find that defenders emphasize short-term performance targets, which areoften based on meeting accounting benchmarks (e.g., return on assets). Ittner et al. (1997)find that defenders place more weight on financial performance measures in chief executiveofficer bonus contracts when compared to prospectors. Thus, defenders’ compensationpackages also provide incentives for managers to misreport financial results because oftheir strong focus on meeting shorter-term accounting performance benchmarks (Ittner etal. 1997; Rajagopalan 1997; Singh and Agrawal 2002). Thus, because prospectors anddefenders both have compensation packages that provide incentives to misreport financial



results, we cannot predict ex ante which strategy would be more likely to misreport withregard to this SAS No. 99 incentive factor.

SAS No. 99 indicates that the “need to obtain additional debt or equity financing tostay competitive—including financing of major research and development [projects]”results in greater incentives for companies to misreport (AU 316.85 [AICPA 2002]).Regarding poor performance, SAS No. 99 indicates that risk factors include operatinglosses and recurring negative cash flows from operations (AU 316.85 [AICPA 2002]).Thus, the performance-related incentive factors highlighted in SAS No. 99 tie directly intothe concept of business risk—i.e., the risk that a company’s economic condition will dete-riorate sometime in the future (Johnstone 2000; Stanley 2011).

Prospectors’ tendency to continually seek new and innovative product and marketopportunities requires them to invest heavily in R&D activities, giving them greater needfor financing but also leaving them vulnerable to overextending their resources andincreasing their risk of incurring losses (Miles and Snow 1978, 2003). On the other hand,defenders engage in minimal R&D activity because they focus on producing a stable andnarrow product line efficiently, where their cost reduction strategy reduces the risk of over-extending their resources and encountering losses (Miles and Snow 1978, 2003). There issome empirical support regarding the lower profitability tendencies of prospectors relativeto defenders. For example, Ittner et al. (1997) find that prospectors are positively associ-ated with financial distress, while Hambrick (1983) finds that prospectors have significantlylower operating cash flow and return on investment (ROI) ratios compared to defenders.Thus, prospectors’ exposure to lower profitability and their need to obtain financing fortheir extensive R&D activities suggest they represent greater business risk and may bemore likely to experience financial reporting irregularities.

Opportunities

SAS No. 99 indicates that opportunity factors that increase the risk of financial reportingirregularities include ineffective monitoring, internal control deficiencies, and the stabilityand complexity of the organizational structure (AICPA 2002). While prior research hasprovided evidence that effective governance mechanisms (e.g., boards of directors, auditcommittees, external auditors) and internal controls are deterrents of financial reportingirregularities (Hogan et al. 2008), a relatively unexplored opportunity risk factor is compa-nies’ organizational stability and complexity.

According to SAS No. 99, firms with an overly complex structure and high turnover ofsenior management or board members are likely to have greater opportunities to engage inmisreporting (AICPA 2002). Prospectors exhibit these characteristics; thus, managers ofthese firms may have more opportunities to misreport. For example, prospectors have a“transitory dominant coalition” (i.e., the tenure of senior management tends to be shorter),while defenders maintain a more stable senior management team (Miles and Snow 1978,2003). Prospectors require decentralized control to facilitate and coordinate their diverseand numerous operations, while defenders require strict, centralized control in their hierar-chical organizational structure. Furthermore, coordination mechanisms in prospectors arecomplex compared to the simple coordination mechanisms in defenders (Miles and Snow1978, 2003). Consistent with the need to maintain flexibility within their organizationalstructure, Simons (1987) finds that prospectors modify their internal control systems muchmore frequently than defenders. Together, these differences suggest prospectors have higherrisk of financial reporting irregularities than defenders because of their decentralized opera-tions and the greater instability and complexity in their organizational structure (Loeb-becke, Eining, and Willingham 1989; AICPA 2002).

In summary, we expect that prospectors are more likely than defenders to engage infinancial reporting irregularities because prospectors demonstrate a greater number of



incentives (e.g., rapid growth, stock-based compensation contracts, greater need to financeextensive R&D activity, tendencies toward lower profitability) and opportunities (e.g.,organizational instability and complexity) to misreport. Stated formally:

HYPOTHESIS 1. Prospector business strategies are more positively associated with financialreporting irregularities than defender business strategies.

Business strategy and audit effort

If organizational business strategies vary in their level of business risk and likelihood offinancial reporting irregularities, we expect audit firms to exert different levels of auditeffort based on their clients’ business strategies.5 Both U.S. and international auditing pro-nouncements (e.g., AS No. 5; ISA 315) emphasize a risk-based approach in planning theaudit (AICPA 2006; IFAC 2009). In particular, ISA 315 emphasizes a business-riskapproach considering factors such as the client’s business strategy (IFAC 2009). In addi-tion, leading audit firms began employing a business risk approach during the 1990s toimprove the efficiency and effectiveness of the audit (Bell, Doogar, and Solomon 2008),and U.S. audit pronouncement AS No. 5 has also encouraged audit firms to take a “moreflexible, top-down risk-based audit approach” (Doogar, Sivadasan, and Solomon 2010:796).

Despite the emphasis on client business strategy as an underlying component of clientbusiness risk in professional guidance, Hay, Knechel, and Wong’s 2006 meta-analysis ofaudit fee models indicates that business strategy is not included as a determinant of auditfees in any study since 1980.6 Rather, Hay et al. (2006) indicate that audit fee researchemploys client risk proxies such as receivable and inventory ratios to capture inherent risk(i.e., accounts more susceptible to material misstatement) and profitability and leverage mea-sures to reflect “the extent to which the auditor may be exposed to loss in the event that aclient is not financially viable” (170).7 These and other proxies for client risk in the audit feeliterature have produced mixed results (Cobbin 2002; Hay et al. 2006). Hay et al. (2006)posit that commonly used proxies for client risk, such as profitability ratios, return on asset(ROA) ratios, and loss, indicators provide mixed results because “auditors may not be asfinely calibrated to differences in the profitability metrics as the fee model suggests” (170).

In our study, we suggest that the business strategies defined by Miles and Snow (1978,2003) represent a more comprehensive approach to addressing client business risk. Thus,consistent with auditing standards (e.g., AS No. 5; ISA 315), a business strategy approachis more likely to reflect auditors’ broader views of client business risk. If auditors considertheir clients’ business strategy as an underlying component of client business risk, weexpect that prospectors require greater audit effort because of their risk-oriented focus,tendencies toward lower profitability, and other risk characteristics (e.g., rapid growth,organizational instability, and complexity). On the other hand, we expect defenders require

5. We do not expect auditors to use the terms prospector or defender in describing their clients’ business strat-

egies. Doing so would require familiarity with the Miles and Snow typology. However, to the extent that

auditors recognize collective firm characteristics that are consistent with the suggested characteristics noted

by Miles and Snow, or the other proposed strategy classifications, audit firms will set engagement fees con-

sistent with differences in these types of clients.

6. In the extant literature, unobservable audit effort is represented by observable audit fees, which is supported

by research linking client-related factors (e.g., size, complexity, and risk) to audit team efficiency in terms

of quantity and mix of labor employed (e.g., Simunic 1980; O’Keefe, Simunic, and Stein 1994; Hackenbrack

and Knechel 1997; Knechel, Rouse, and Schelleman 2009).

7. Although Hay et al. (2006) do not classify the latter risk, Stanley (2011) provides a discussion of how a cli-

ent’s susceptibility to loss affects both client and auditor business risk. Furthermore, Johnstone (2000) pro-

vides empirical evidence that suggests that client business risk directly impacts the auditor’s business risk

assessment.



lower audit effort (lower audit fees) because they exhibit fewer risk characteristics (e.g.,less risk-oriented focus, tendencies toward profitability, cautious and incremental growthpatterns, organizational stability, and less complexity) relative to prospectors. Thus, whencompared to defenders, prospectors’ unique client business risks likely result in auditorsexerting greater audit effort for prospectors in order to reduce audit risk to an acceptablelevel. Stated formally:

HYPOTHESIS 2. Prospector business strategies are more positively associated with the levelof audit effort than defender business strategies.

4. Measures and models

Business strategy composite measure

Relying on Miles and Snow (1978, 2003), we construct a discrete STRATEGY compositemeasure, which proxies for the organization’s business strategy. Higher STRATEGYscores represent companies with prospector strategies and lower scores represent compa-nies with defender strategies. Similar to Ittner et al. 1997, we use the following characteris-tics for the STRATEGY composite measure: (a) the ratio of research and development tosales, (b) the ratio of employees to sales, and (c) a historical growth measure (one-yearpercentage change in total sales).8 Ittner et al. (1997) used a fourth measure, the numberof new product or service introductions, which requires access to a proprietary database.As a substitute for this measure using publicly available data, we use (4) the ratio ofmarketing (SG&A) to sales, which Hambrick (1983) found empirically differentiatesprospectors from defenders. Miles and Snow’s (1978, 2003) theory suggests that otherimportant attributes that distinguish prospectors and defenders include organizationalstability regarding the length of employee tenure and the efficiency and automation ofoperations as reflected in overall capital intensity.9 We capture these two attributes using(5) a measure of employee fluctuations (standard deviation of total employees) and (6) ameasure of capital intensity (net PPE scaled by total assets), respectively.

Consistent with Ittner et al. 1997, all variables are computed using a rolling averageover the prior five years. Each of the six individual variables is ranked by forming quin-tiles within each two-digit SIC industry-year. Within each company-year, those observa-tions with variables in the highest quintile are given a score of 5, in the second-highestquintile are given a score of 4, and so on, and those observations with variables in thelowest quintile are given a score of 1. Then for each company-year, we sum the scoresacross the six variables such that a company could receive a maximum score of 30 (pros-pector-type) and a minimum score of 6 (defender-type). Refer to Appendix 2 for detailsrelated to our business strategy composite measure construction and Appendix 3 forexamples of companies in each strategy.10

8. Ittner et al. (1997) use market-to-book ratio to proxy for growth. We use the 1-year percentage change in

sales for the growth component of our strategy measure and include book-to-market as a control variable

in our models of financial reporting irregularities. In sensitivity tests, we replace our growth proxy with

market-to-book following Ittner et al. 1997 and our results are robust in our audit fee model and two of

our irregularity samples (AAERs and lawsuits).

9. Hambrick (1983) provides empirical evidence that relative to prospectors, defenders are more automated

and efficient (i.e., defenders have higher ratios of gross assets to employees, higher value added to employ-

ees, and lower direct costs) and concludes that defenders are more capital intensive while prospectors have

“more flexible, labor intensive capacity configurations” (23).

10. We also use factor scores as an alternative to our composite measure of business strategy and obtain simi-

lar results. We discuss the results of this additional test in greater detail in section 7.



Financial reporting irregularities

We use logistic regression to determine whether business strategy is associated with ourthree financial reporting irregularities: AAERs, lawsuits due to accounting improprieties,and accounting restatements. Our model for the likelihood of financial reporting irregular-ities is as follows with subscripts omitted:

FR IRREGULARITY ¼ b0 þ b1STRATEGYþ b2lnðASSETSÞ þ b3ROAþ b4LOSS

þ b5BTM þ b6SALES GROWTHþ b7M&Aþ b8LEVERAGE

þ b9FINANCINGþ b10FIRMAGEþ b11HERFþ b12LITIGIOUS

þ b13DAPþ b14BIGNþ b15DED IOþ Industry FixedEffects

þ Year FixedEffectsþ e ð1Þ:

For each irregularity model, we define the dependent variable based on the periodsduring which misreporting occurred. Thus, FR_IRREGULARITY takes a value of 1 forperiods misreported according to the AAER, lawsuit, or restatement, and zero otherwise,in each of the respective models.11 The remaining variables are defined in Table 1.

Based on Hypothesis 1, we expect a positive sign for the coefficient on STRATEGY.For the remaining coefficients we discuss them in the context of SAS No. 99, and indicatewhether the variables represent the incentive or opportunity to engage in financial report-ing irregularities. We include the following variables to proxy for incentives: firm size (ln(ASSETS))12; profitability (ROA; LOSS); growth (BTM; SALES GROWTH); mergerand acquisition activity (M&A); leverage (LEVERAGE); financing needs (FINANC-ING); and firm age (FIRM AGE). We do not predict a sign for the coefficients on ln(ASSETS) or LEVERAGE based on the mixed results in prior research (e.g., Beasley,Carcello, and Hermanson 1999; Turner and Weirich 2006; McGuire, Omer, and Sharp2012). Based on prior research (Dechow, Sloan, and Sweeney 1996; Beneish 1997, 1999;Summers and Sweeney 1998; Beasley et al. 1999; Erickson, Hanlon, and Maydew 2006;Dechow, Ge, Larson, and Sloan 2011; McGuire et al. 2012), we expect a negative coeffi-cient on ROA and a positive coefficient on LOSS; we expect a negative coefficient forBTM and do not predict direction for the coefficient on SALES GROWTH; we expect apositive coefficient on M&A and a positive coefficient on FINANCING; we expect a nega-tive coefficient on FIRM AGE. We control for industry competition using a HerfindahlIndex (HERF) and expect a positive coefficient because higher values of HERF indicategreater industry competition, which is an incentive risk factor from the SAS No. 99 frame-work (AICPA 2002).

The factors that represent significant opportunity risk are industry conditions,accounts based on significant estimates or unusual transactions, and ineffective monitoring(AICPA 2002). We follow prior research (e.g., Beneish 1999; Farber 2005; Hogan et al.2008; Dechow et al. 2011) to determine our variables. We include an indicator variable forlitigious industries (LITIGIOUS) and expect a positive coefficient because these industrieshave a greater likelihood of lawsuits (Francis, Philbrick, and Schipper 1994). We includediscretionary accruals (DAP) because prior research suggests high or unusual accrualsindicate high-risk audits (e.g., Beneish 1997, 1999; Dechow et al. 2011). However, Schell-eman and Knechel (2010) argue that auditors are presumably aware of their clients’accounting choices and thus the level of accruals may not indicate high risk. Therefore, we

11. Although our main model tests each type of irregularity separately, in later sensitivity analyses we combine

the irregularities to form a single irregularity indicator and examine the impact of business strategy in the

combined sample.

12. Our results are consistent if we control for size using the natural log of market value of equity.



TABLE 1

Variable descriptions

Variable Description

STRATEGY = Discrete score with values ranging from six to 30 where high (middle)[low] values indicate prospector (analyzer) [defender] firms, respectively;

refer to Appendix 2 for details about the score construction.RDS5 = Ratio of research and development expenditures to sales computed over

a rolling five-year average.

EMPS5 = Ratio of the number of employees to sales computed over a rolling priorfive-year average.

REV5 = One-year percentage change in total sales computed over a rolling priorfive-year average.

SGA5 = Ratio of selling, general and administrative (SG&A) expenses to salescomputed over a rolling prior five-year average.

r(EMP5) = Standard deviation of the total number of employees computed over

a rolling prior five-year period.CAP5 = Capital intensity measured as net property, plant, and equipment scaled

by total assets and computed over a rolling prior five-year average.

AAER = Indicator variable equal to 1 if the firm experiences an AAER (obtainedfrom Audit Integrity, a proprietary source) anytime during the yearand 0 otherwise.

LAWSUIT = Indicator variable equal to 1 if the firm experiences a lawsuit due to an

accounting irregularity (obtained from Audit Analytics) anytimeduring the year and 0 otherwise.

RESTATE = Indicator variable equal to 1 if the firm experiences a restatement

during the year and 0 otherwise.ln(ASSETS) = Natural logarithm of total assets.ROA = Return on assets equal to income before extraordinary items divided

by total assets.LOSS = Indicator variable equal to 1 if a loss occurred within the current

or previous two fiscal years (income before extraordinary items isnegative) and 0 otherwise.

BTM = Book-to-market ratio equal to total common equity outstandingdivided by the market capitalization at the end of the fiscal year.

SALES GROWTH = Percentage change in sales from the prior year to the current year.

M&A = Indicator variable equal to 1 if a merger or acquisition occurred inprior five years and 0 otherwise from COMPUSTAT footnote codes.

LEVERAGE = Financial leverage equal to total debt divided by total assets.

FINANCING = Ex ante measure of firm’s desire for financing (based on Dechow et al.1996; Erickson et al. 2006) where an indicator variable is equal to 1if the firm’s free cash is less than �0.5 and 0 otherwise. WhereFreeCash = {[Cash from operationst – Avg Capital

Expenditures t-3 to t-1] / Current Assetst-1}.FIRM AGE = Length of time in years the firm has been publicly listed based on

the company’s initial public offering (IPO) date from the Center for

Research in Security Prices (CRSP).HERF = Herfindahl Index (based on Cheng 2005) equal to the sum of squares of

market shares of all companies in an industry (three-digit SIC) where

market share equals the company with sales in time t divided by sumof (sales)t over the three-digit SIC industry, where the higher theHerfindahl Index, the more concentrated the industry.

(The table is continued on the next page.)



offer no prediction for the coefficient on DAP.13 We include an indicator for Big N auditfirms (BIGN) and use dedicated institutional ownership (DED IO) to represent externalmonitoring (e.g., Farber 2005; McGuire et al. 2012). We expect a negative coefficient forBIGN. However, Farber (2005) finds no difference in the level of institutional investors forsamples of fraud and non-fraud firms; thus, we make no prediction for the coefficient onDED IO.14 To control for potential firm, industry, and time fixed effects in our panel data,we include year and industry indicators and cluster by firm (Petersen 2009).

TABLE 1 (Continued)

Variable Description

LITIGIOUS = Indicator variable equal to one if the company is in a litigious industry

as defined by Ali and Kallapur 2001 (based on Francis et al. 1994),where the following industries are litigious: SIC 2833–2836, 3570–3577,3600–3674, 5200–5961, 7370–7374, and 8731–8734.

DAP = Discretionary accruals, which is the residual of the following

performance-adjusted modified Jones model (based on Larcker andRichardson 2004), estimated by 2-digit SIC industry every year: whereTotal

Accruals = a + b1(ΔSALES - ΔREC) +b2PPE + b3BTM +b4CFO + e.BIGN = Indicator variable equal to 1 if a Big N auditor is used and 0

otherwise.

DED IO = Lagged value of a dedicated institutional investor variable followingBushee and Noe 2000 and Bushee 2001 from Thomson Financial.

AUDIT FEE (MM$) = Total audit fees from Audit Analytics in millions of dollars.ln(AUDIT FEE) = Natural logarithm of total audit fees from Audit Analytics.

sqrt(BUS SEG) = Square root of the number of business segments from COMPUSTATsegment file.

FOREIGN = Percentage of foreign sales from COMPUSTAT segment file.

RECINV = Sum of receivables and inventories scaled by total assets.QUICK = Quick ratio equals total current assets less inventories to total current

liabilities.

MW = Material weakness indicator variable equal to 1 if the company hadan internal control weakness attributed to Sarbanes-Oxley Act sections:302 or 404 from Audit Analytics and 0 otherwise.

SPECIALIST = Industry specialist variable following Francis et al. 2005 in defining

a metropolitan statistical area (MSA) level industry specialist.AUD TENURE = Auditor tenure in years from COMPUSTAT.YE = Busy season indicator variable equal to 1 if the company’s fiscal

year end is December 31 and 0 otherwise.AUD OPIN = Indicator variable equal to 1 if the audit opinion was anything

other than a standard, unqualified opinion and 0 otherwise.

ln(NON AUDIT) = Natural logarithm of total nonaudit fees from Audit Analytics.

13. In untabulated robustness tests, we replace discretionary accruals with Dechow and Dichev’s 2002 accruals

quality measure because Jones, Krishnan, and Melendrez (2008) and Price, Sharp, and Wood (2011) find

the Dechow and Dichev 2002 measure is more strongly associated with financial reporting irregularities.

We note that while accruals quality is positive and significant in our AAER and lawsuit models, the coeffi-

cient on STRATEGY remains positive and significant in all three models.

14. We rely on dedicated institutional investors as a monitoring variable rather than other measures of gover-

nance such as board of director or audit committee characteristics because of concerns about a potentially

endogenous relation between these alternative governance characteristics and our STRATEGY variable.



Audit fee model

We construct an audit fee model based on standard measures from the audit fee literatureand rely on Hay et al.’s 2006 framework to classify our audit fee model controls accordingto each major audit fee determinant (refer to Table 2 in Hay et al. 2006). Our model is asfollows, with subscripts omitted:

lnðAUDIT FEEÞ ¼ b0 þ b1STRATEGYþ b2lnðASSETSÞ þ b3sqrtðBUS SEGÞþ b4FOREIGN þ b5RECINVþ b6ROAþ b7LOSSþ b8LEVERAGE

þ b9QUICKþ b10MWþ b11DED IOþ b12BIGNþ b13SPECIALIST

þ b14AUD TENURE þ b15YEþ b16AUD OPINþ b17lnðNON AUDITÞþ Industry Fixed Effectsþ Year Fixed Effectsþ e ð2Þ:

We provide variable definitions in Table 1. We discuss our expectations for the modelcoefficients below, based on Miles and Snow (1978, 2003) and the prior audit fee litera-ture. Based on Hypothesis 2, we expect the coefficient for STRATEGY to be positive ifauditors exert greater effort for prospectors and lower effort for defenders. The primaryclient attributes that explain audit fees are client size, complexity, and risk (Hay et al.2006). Client size is measured using the natural logarithm of total assets (ln(ASSETS)),and based on prior research, we expect a positive coefficient on ln(ASSETS). We proxyfor client complexity using the number of business segments (sqrt(BUS SEG)) and thepercentage of foreign-based sales (FOREIGN).15 Similar to prior research, we expect posi-tive coefficients for our measures of client complexity (Casterella, Francis, Lewis, andWalker 2004; Larcker and Richardson 2004; Francis, Reichelt, and Wang 2005; Hay et al.2006; Hogan and Wilkins 2008; Hribar, Kravet, and Wilson 2009).

Based on Hay et al. (2006), we also include in the model proxies associated with risks rel-evant to an audit engagement: factors that increase inherent risk and factors that increaseauditor business risk. Inherent risk factors relate to the difficulty and greater risk of error inauditing certain accounts (e.g., receivables or inventories), while auditor business risk factorsrelate to the risk of managing an audit firm and the auditor’s exposure to potential loss (Hayet al. 2006; Stanley 2011). We argue that business strategy likely represents the risks associ-ated with specific client strategy choices; thus, we include these additional risk proxies to sep-arate the effects of risk related to engaging in audit practice from client business risk. Thisseparation is consistent with Stanley 2011 who separately identifies the risks that are specificto offering services as an auditor versus the risks specific to the client business operation. Weproxy for inherent risk using the proportion of receivables and inventories in total assets(RECINV), and we expect a positive coefficient on RECINV. We proxy for auditor businessrisk using return on assets (ROA) and a loss indicator (LOSS), and we expect a negativecoefficient on ROA and a positive coefficient on LOSS (Hay et al. 2006). To capture addi-tional proxies related to auditor business risk, we use two common leverage ratios, the ratioof total debt to assets (LEVERAGE) and the quick ratio (QUICK) (Hay et al. 2006).

Hay et al. (2006) also suggest internal control weaknesses, governance or monitoringmechanisms, and industry characteristics as client risk factors that should be included inan audit fee model. The effectiveness of internal controls is represented in our model bymaterial weakness occurrences (MW), and we expect a positive coefficient on MW

15. We follow Francis et al. 2005 in using business segments and percentage of foreign-based sales for com-

plexity. Potential alternative measures of complexity include the number of subsidiaries or the proportion

of foreign subsidiaries; however, the number of subsidiaries is not readily available for U.S. firms. Thus,

we rely on the number of business segments as our proxy for complexity. In robustness tests, we replace

the percentage of foreign-based sales with percentage of foreign-based assets, and our audit fee results are

robust.



(Raghunandan and Rama 2006; Hogan and Wilkins 2008). To control for external moni-toring, we use dedicated institutional investors (DED IO) (Bushee and Noe 2000; Bushee2001). Based on Hay et al. (2006), we expect a positive coefficient on DED IO. We includeindustry fixed effects in the model to control for industry characteristics.

Hay et al. (2006) indicate that auditor attributes such as auditor quality and auditortenure affect audit fees. Both Big N audit firms (BIGN) and industry specialists (SPE-CIALIST) are typically associated with higher audit quality; thus, we expect positive coef-ficients for BIGN and SPECIALIST (e.g., DeAngelo 1981; Palmrose 1988; Francis et al.2005; Hay et al. 2006; Reichelt and Wang 2010). Although prior research suggests an asso-ciation between auditor tenure (AUD TENURE) and audit fees (Casterella et al. 2004;Hay et al. 2006; Hribar et al. 2009), Hay et al.’s 2006 meta-analysis indicates mixedresults; therefore, we do not predict a sign for the coefficient on AUD TENURE.

Engagement attributes such as busy season, audit problems, and nonaudit services alsoaffect audit fees (Hay et al. 2006). We measure busy season using a calendar year-endreporting indicator variable (YE), and we expect a positive coefficient based on Hay etal.’s 2006 meta-analysis. Additionally, we expect a positive coefficient on both modifiedaudit opinions (AUD OPINION) and non-audit services (ln(NON AUDIT)). Finally, to

TABLE 2

Sample selection

Description Company-years

Panel A: Strategy composite score construction

COMPUSTAT data for years between 1980 and 2009(zero negative sales and assets and missing historical SIC codes removed)

178,388

Less Utilities and Financial Industries (SIC 4900–99 and 6000–99) (33,628)Less required 5-year rolling average data for STRATEGY measure (70,404)Less missing values for all 6 STRATEGY component variablesper company-year

(16,839)

Total observations for STRATEGY composite score

data set (1993–2009)57,517

Panel B: Financial reporting irregularities samples

STRATEGY composite score dataset in panel A 57,517Less missing control variable values for irregularity models (41,227)

Total observations for STRATEGY-irregularity control variable data set 16,290

Merge AAER-COMPUSTAT data set (1,257 company-years or 406 uniqueAAERs) with STRATEGY-irregularity control variable data set for 206

AAER-years (1993–2006)

11,679

Merge Lawsuit-COMPUSTAT data set (2,716 company-years or1,205 unique lawsuits) with STRATEGY-irregularity control variable data

set for 344 lawsuit-years (1995–2009)

15,449

Merge Restatement-COMPUSTAT data set (9,172 company-years or4,625 unique irregular restatements) with STRATEGY-irregularity controlvariable data set for 1,763 restatement-years (1998–2009)

14,409

Panel C: Audit fee model sample

STRATEGY composite score data set in panel A 57,517

Less missing audit fee data and control variables in audit fee model (45,680)Total observations for STRATEGY audit fee data set (2001–2009): 11,837



control for potential firm and time fixed effects in our panel data we include year indica-tors and cluster by firm (Petersen 2009).16

5. Data

Table 2 outlines our sample selection process. We construct our business STRATEGYmeasure using all firms from the COMPUSTAT Annual file for fiscal years between 1980and 2009 with positive sales and asset observations, and non-missing historical SICcodes.17 We then delete utilities and financial industries (SIC 4900–99 and 6000–999) dueto the regulated nature of these industries. All data used to construct the STRATEGYmeasure requires a five-year rolling average (e.g., Ittner et al. 1997).18 After imposing thefive-year rolling average constraint and deleting missing data, we have a sample of 57,517company-year observations from 1993 to 2009 for the STRATEGY measure.

We construct our financial reporting irregularities samples from three data sources: (a)a sample of SEC AAERs, obtained from Audit Integrity; (b) shareholder lawsuits relatedto accounting improprieties, obtained from the Audit Analytics litigation database; and(c) accounting restatements, obtained from the Audit Analytics restatements database.Our AAER sample contains 11,679 company-year observations with 206 AAER-yearobservations between fiscal years 1993 to 2006; our lawsuit sample contains 15,449 com-pany-year observations with 344 lawsuit-year observations between fiscal years 1995 to2009; and our restatement sample contains 14,409 company-year observations with 1,763restatement-year observations between fiscal years 1998 to 2009. Because some of theirregularities overlap, our combined sample includes 2,083 company years associated withfinancial reporting irregularities, which we examine in a combined irregularity sample inlater robustness tests.

Our audit fee sample is based on fiscal years 2001 to 2009, using audit fee data fromAudit Analytics and all available COMPUSTAT data for the control variables in theaudit fee model. Our audit fee sample contains 11,837 company-year observations.

The descriptive statistics for our samples are presented in Table 3; we also present sep-arate descriptive statistics for companies following prospector and defender strategies.Table 3, panel A presents the industry affiliations for our sample, where 57,517 total com-pany-year observations represent 4,112 prospector-years and 3,473 defender-years, and theremaining observations representing our analyzer category.19 Consistent with expectations,

16. Although we are consistent with the audit fee literature in measuring both our control variables and our

dependent variable (ln(AUDIT FEES)) in time t, to the extent that audit fees are determined at the begin-

ning of the client’s fiscal year, our dependent variable (ln(AUDIT FEES)) could potentially be determined

before our independent variables. However, this timing concern is less problematic for our variable of

interest, STRATEGY, because this variable is computed using a rolling prior 5-year average. To address

this potential timing concern, in robustness tests we take a lead value (t + 1) of our dependent variable,

ln(AUDIT FEES), and reestimate our audit fee model using time t for our explanatory variables. This

method ensures that our dependent variable is determined after the explanatory variables. Our results

remain robust to this specification.

17. Based on prior research by Jan and Ou 2008, we do not exclude negative book-to-market companies.

Because companies with significant R&D expenditures may represent negative book value companies and

R&D expenditures are an input into our STRATEGY measure, we would omit a large number of potential

prospector companies and impose a selection bias on the sample if we deleted negative book value compa-

nies. However, if we delete negative book-to-market firms, our inferences are unchanged.

18. We require at least three years of non-missing observations for each of our measures to preserve observa-

tions, provided that the company has at least six years of consecutive data in COMPUSTAT. When we

either tighten or ease this restriction, our inferences remain unchanged.

19. Because some firm-year observations in our sample have STRATEGY scores just below (above) the cutoff

point for prospectors (defenders), our sample of analyzers is also likely to include some firms that are “less

obvious” prospectors or defenders. For simplicity, we refer to all of the firms between our two cutoff

points as analyzers.



TABLE 3

Descriptive statistics

Panel A: Industry affiliations (company-years)

Two-digitSIC code Industry affiliation

Full sample(N = 57,517)

Prospectors(N = 4,112)

Defenders(N = 3,473)

Number Percent Number Percent Number Percent

01–09 Agriculture, Forestryand Fishing

252 0.4 16 0.4 14 0.4

10–14 Mining 2,967 5.2 250 6.1 126 3.615–17 Construction 720 1.3 34 0.8 35 1.020–39 Manufacturing 32,103 55.8 2,557 62.2 2,142 61.7

40–48 Transportation andCommunicationsServices

2,869 5.0 203 4.9 174 5.0

50–51 Wholesale Trade 2,927 5.1 189 4.6 147 4.252–59 Retail Trade 5,040 8.8 202 4.9 161 4.670–89 Services 10,540 18.3 652 15.9 666 19.299 Other 99 0.2 9 0.2 8 0.2

Total 57,517 100.0 4,112 100.0 3,473 100.0

Panel B: Composite and component STRATEGY (means and medians in bold are significantlydifferent at p < 0.05)

Variable



Defenders(N = 3,473)

Mean Med Q1 Q3 Std dev Mean Med Mean Med

STRATEGY 18.04 18.00 16.00 20.00 3.63 25.09 25.00 10.96 11.00

STRATEGY component variables:

RDS5 0.18 0.00 0.00 0.06 6.07 1.34 0.14 0.01 0.00

EMPS5 0.01 0.01 0.00 0.01 0.06 0.03 0.01 0.00 0.00

REV5 61.77 10.34 2.67 23.10 1,966.62 316.10 43.27 5.17 2.62

SGA5 0.69 0.25 0.15 0.41 13.21 3.74 0.63 0.16 0.14

r(EMP5) 1.38 0.17 0.04 0.78 5.75 1.34 0.13 0.26 0.05

CAP5 0.28 0.22 0.11 0.40 0.22 0.17 0.12 0.43 0.41

Panel C: Financial reporting irregularity model variables (means and medians in bold aresignificantly different at p < 0.05)

Variable



Defenders(N = 880)


AAER 0.01 0.00 0.00 0.00 0.11 0.03 0.00 0.01 0.00

LAWSUIT 0.02 0.00 0.00 0.00 0.14 0.04 0.00 0.01 0.00

RESTATE 0.11 0.00 0.00 0.00 0.31 0.12 0.00 0.07 0.00

STRATEGY 18.43 18.00 16.00 21.00 3.71 25.18 25.00 11.03 11.00

ln(ASSETS) 5.50 5.52 4.16 6.78 1.98 5.02 4.98 5.31 5.25

ROA �0.06 0.03 �0.05 0.08 0.58 �0.20 �0.03 �0.01 0.02




all three viable business strategies (prospectors, defenders, and analyzers) are common ineach industry (Miles and Snow 1978, 2003). We find that the percentages of prospectorsand defenders in each industry are similar to each other and to the percentage of the totalsample in each industry. For example, the largest industry segment represented in oursample is the manufacturing sector (two-digit SIC codes 20–39), which consists of approxi-

TABLE 3 (Continued)

Variable

Full sample

(N = 16,290)

Prospectors

(N = 1,505)

Defenders

(N = 880)


LOSS 0.52 1.00 0.00 1.00 0.50 0.72 1.00 0.51 1.00

BTM 0.50 0.46 0.25 0.79 1.44 0.37 0.29 0.62 0.57

SALES GROWTH 14.36 8.00 �3.91 21.46 128.70 28.73 14.71 5.33 3.51

M&A 0.55 1.00 0.00 1.00 0.50 0.63 1.00 0.34 0.00

LEVERAGE 0.23 0.15 0.01 0.34 0.44 0.22 0.10 0.29 0.23

FINANCING 0.06 0.00 0.00 0.00 0.24 0.16 0.00 0.05 0.00

FIRM AGE 10.91 9.59 7.01 13.07 6.35 9.55 8.18 11.33 10.47

HERF 0.15 0.10 0.06 0.18 0.15 0.14 0.08 0.20 0.15

LITIGIOUS 0.41 0.00 0.00 1.00 0.49 0.50 1.00 0.24 0.00

DAP 0.04 0.04 �0.02 0.11 0.14 0.03 0.04 0.06 0.05

BIGN 0.79 1.00 1.00 1.00 0.41 0.79 1.00 0.78 1.00

DED IO 0.07 0.03 0.00 0.11 0.08 0.06 0.02 0.07 0.03

Panel D: Audit fee model variables (means and medians in bold are significantly different atp < 0.05)

Variable

Full sample (N = 11,837)Prospectors(N = 754)

Defenders(N = 779)


AUDIT FEE (MM$) 1.62 0.76 0.28 1.76 2.53 1.51 0.79 1.01 0.48

ln(AUDIT FEE) 13.49 13.53 12.53 14.38 1.29 13.52 13.58 13.05 13.07

STRATEGY 17.78 18.00 15.00 20.00 3.59 25.10 25.00 11.02 11.00

ln(ASSETS) 5.92 5.97 4.51 7.31 2.01 5.51 5.50 5.48 5.67sqrt(BUS SEG) 1.42 1.00 1.00 1.73 0.48 1.35 1.00 1.43 1.41

FOREIGN 0.24 0.15 0.00 0.43 0.27 0.29 0.23 0.18 0.03

RECINV 0.29 0.27 0.15 0.40 0.18 0.22 0.19 0.32 0.31

ROA �0.01 0.04 �0.02 0.08 0.20 �0.15 �0.01 0.01 0.04

LOSS 0.47 0.00 0.00 1.00 0.50 0.68 1.00 0.44 0.00

LEVERAGE 0.20 0.16 0.01 0.31 0.22 0.21 0.14 0.24 0.21

QUICK 2.04 1.47 0.97 2.47 1.77 2.59 1.81 1.74 1.33

MW 0.12 0.00 0.00 0.00 0.32 0.15 0.00 0.09 0.00

DED IO 0.08 0.06 0.00 0.13 0.08 0.07 0.04 0.07 0.05BIGN 0.77 1.00 1.00 1.00 0.42 0.75 1.00 0.71 1.00SPECIALIST 0.51 1.00 0.00 1.00 0.50 0.48 0.00 0.48 0.00

AUD TENURE 8.93 7.00 3.00 12.00 8.32 8.01 6.00 8.38 6.00YE 0.64 1.00 0.00 1.00 0.48 0.68 1.00 0.64 1.00AUD OPIN 0.54 1.00 0.00 1.00 0.50 0.53 1.00 0.49 0.00

ln(NON AUDIT) 11.84 11.85 10.71 13.00 1.67 11.75 11.78 11.51 11.49

Notes:

Variable definitions are provided in Table 1.



mately 56 percent of the total firms. Similarly, the manufacturing sector remains the larg-est industry segment represented within each of the prospector and defender firm subsam-ples, consisting of approximately 62 percent of firms in each subsample.

Table 3, panel B provides descriptive statistics for our STRATEGY composite mea-sure and the raw components of STRATEGY, while panels C and D provide descriptivestatistics for variables in our financial reporting irregularity and audit fee models, respec-tively. Consistent with expectations, prospectors and defenders have significantly differentmeans and medians (p < 0.05) for each of the six STRATEGY components—i.e., R&D/sales (RDS5), employees/sales (EMPS5), percentage change in total revenue (REV5),SG&A/sales (SGA5), standard deviation of total number of employees [r(EMP5)], andcapital intensity (CAP5).

In the financial reporting irregularity samples (Table 3, panel C), prospectors have sig-nificantly higher occurrences of AAERs, lawsuits, and restatements compared to defend-ers, consistent with Hypothesis 1. Prospectors and defenders show significant differences inmean and median values (p < 0.05) across all control variables except for BIGN and DEDIO. Compared to defenders, prospectors are smaller (ln(ASSETS)), less profitable (ROA;LOSS), more growth-oriented (BTM; SALES GROWTH), engage in more mergers andacquisitions (M&A), have lower leverage (LEVERAGE), and have a greater need forfinancing (FINANCING). Prospectors are also younger (FIRM AGE), involved in less con-centrated markets (HERF), operate in more litigious industries (LITIGIOUS), and havelower discretionary accruals (DAP).

In the audit fee sample (Table 3, panel D), prospectors have higher audit fees (ln(AUDIT FEE)), consistent with Hypothesis 2. There are no significant differences at themean or median (p > 0.05) between prospectors and defenders for the following variables:ln(ASSETS), DED IO, BIGN, SPECIALIST, AUD TENURE, YE, and AUD OPIN.20

There are significant differences between prospectors and defenders on business segments(sqrt(BUS SEG) and foreign sales (FOREIGN); however, prospectors have higher foreignsales but fewer business segments, which partially supports the Miles and Snow (1978,2003) expectation that prospectors have a more diverse and complex organization. Pros-pectors also have lower inventory quantities than defenders (RECINV) and correspond-ingly have greater short-term liquidity (QUICK). Consistent with prior research (Milesand Snow 1978, 2003; Hambrick 1983; Ittner et al. 1997), prospectors have lower profit-ability (ROA) and greater loss (LOSS) occurrences than defenders. Finally, prospectorshave lower leverage (LEVERAGE), experience more material weaknesses (MW), and payhigher nonaudit fees (ln(NON AUDIT)) than defenders.

In untabulated results, we consider the correlations among STRATEGY and the vari-ables used in our financial reporting irregularities and audit fee models. Our STRATEGYmeasure is positively correlated (p < 0.01) with all three types of financial reporting irregu-larities (AAERs, lawsuits, and restatements), and the level of audit fees. Among the modelvariables, the correlations are consistent with prior literature, and only seven exhibit corre-lations greater than 0.50 including the correlations between audit and nonaudit fees, sizeand audit fees, and size and the use of a Big N auditor.

6. Results

Financial reporting irregularities

Table 4, panel A presents results from estimating our financial reporting irregularitiesmodels. Consistent with Hypothesis 1, STRATEGY is positive and significant in the

20. All the continuous measures in the audit fee model are winsorized at the 1st and 99th percentiles. In the

financial reporting irregularity models, we winsorize only book-to-market and discretionary accruals. How-

ever, our results in the irregularities models are robust to winsorizing all continuous variables.



TABLE

4

Financialreportingirregularities

andauditfeemodel

estimation

Panel

A:Financialreportingirregularities

FR

IRREGULARIT

Y¼

b 0þb1STRATEGYþb2lnðA

SSETSÞþ

b3ROAþb 4LOSSþb5BTM

þb 6SALESGROWTH

þb7M&Aþb8LEVERAGEþb9FIN

ANCIN

Gþb10FIR

MAGE

þb 1

1HERFþb12LIT

IGIO

USþb13DAPþb14BIG

Nþb15DED

IO

þIndustry

Fixed

EffectsþYearFixed

Effectsþe

Dependent

variable:

Sign

AAER

LAWSUIT

RESTATEMENT

Intercept

?�9

.332(�

7.96)***

�11.201(�

7.27)***

�7.575(�

11.45)***

�8.747(�

9.51)***

�4.959(�

15.08)***

�5.539(�

13.93)***

STRATEGY

+0.100(2.05)**

0.062(2.29)**

0.032(2.69)***

ln(ASSETS)

?0.307(3.79)***

0.301(3.59)***

0.355(7.62)***

0.354(7.53)***

0.064(2.26)**

0.062(2.19)**

ROA

��0

.184(�

3.87)***

�0.167(�

3.45)***

�0.159(�

2.84)***

�0.149(�

2.63)***

0.088(1.10)

0.115(1.45)

LOSS

+0.339(1.35)*

0.270(1.05)

0.238(1.43)*

0.203(1.23)

0.300(3.45)***

0.287(3.29)***

BTM

�0.043(0.64)

0.075(0.94)

0.001(0.04)

0.016(0.47)

0.025(1.12)

0.031(1.37)

SALES

GROWTH

?0.000(0.87)

0.000(0.37)

0.000(0.97)

0.000(0.62)

�0.000(�

1.11)

�0.000(�

1.29)

M&A

+0.751(2.82)***

0.621(2.30)**

0.308(1.69)**

0.232(1.23)

�0.041(�

0.48)

�0.078(�

0.90)

LEVERAGE

?�1

.002(�

1.67)*

�0.808(�

1.38)

�0.565(�

1.51)

�0.468(�

1.29)

0.044(0.47)

0.077(0.91)

FIN

ANCIN

G+

�0.375( �

0.62)

�0.533(�

0.87)

0.574(2.11)**

0.455(1.61)*

0.068(0.46)

0.022(0.15)

FIR

MAGE

��0

.003(0.16)

0.004(0.19)

�0.074(�

3.76)***

�0.069(�

3.54)***

�0.020(�

2.43)***

�0.017(�

2.15)**

HERF

+1.161(1.86)**

1.225(2.00)**

1.470(2.56)***

1.460(2.53)***

0.429(1.30)*

0.450(1.37)*

LIT

IGIO

US

+0.288(0.80)

0.235(0.67)

0.931(3.83)***

0.884(3.63)***

0.160(1.44)*

0.140(1.26)

DAP

?�0

.009(�

0.01)

0.250(0.27)

�0.608(�

1.34)

�0.466(�

1.01)

�0.605(�

2.55)**

�0.552(�

2.32)**

BIG

N�

�0.050(�

0.12)

�0.090(�

0.22)

�0.811(�

3.67)***

�0.847(�

3.83)***

0.139(1.20)

0.131(1.14)

DED

IO?

0.791(0.60)

0.884(0.66)

1.921(2.21)**

1.990(2.26)**

1.563(3.29)***

1.584(3.33)***

Industry

&yearfixed

effects

Yes

Yes

Yes

Yes

Yes

Yes

(Thetable

iscontinued

onthenextpage.)



TABLE

4(C

ontinued)

Dependent

variable:

Sign

AAER

LAWSUIT

RESTATEMENT

Observations

11,679

11,679

15,449

15,449

14,409

14,409

McF

adden’s

(pseudo)R2

0.102

0.112

0.121

0.125

0.102

0.104

ROC

curve

0.78

0.78

0.80

0.80

0.73

0.73

Notes:

Coeffi

cientvalues

(z-statistics)

are

shownwithstandard

errors

clustered

atthecompanylevel.**

*,**,*

indicate

statisticalsignificance

atthe1percent,5

percent,and10percentlevels,respectively,tw

o-tailed

(one-tailed

ifpredicted).Thedependentvariable

isanindicatorvariable

equalto

oneifthe

companymisreported

itsfinancialresultsduringthecurrentyearaccordingto

anAAER,lawsuit,orrestatement,andzero

otherwise.

Industry

fixed

effects

use

separate

indicators

forthe17industry

groupingsdefined

inFamaandFrench

1988.Refer

toTable

1forvariable

definitions.

Panel

B:Auditfeemodel

lnðA

UDIT

FEEÞ¼

b0þb1STRATEGYþb2lnðA

SSETSÞþ

b 3sqrtðB

USSEGÞþ

b4FOREIG

N

þb 5RECIN

Vþb6ROAþb7LOSSþb8LEVERAGEþb9QUIC

Kþb10MW

þb11DED

IO

þb 1

2BIG

Nþb13SPECIA

LIS

Tþb14AUD

TENUREþb15YEþb16AUD

OPIN

þb 1

7lnðN

ON

AUDIT

ÞþIndustry

Fixed

EffectsþYearFixed

Effectsþe

Dependentvariable

Sign

ln(AUDIT

FEE)

Intercept

?8.757(88.82)***

8.501(84.60)***

STRATEGY

+0.018(7.63)***

ln(ASSETS)

+0.448(59.86)***

0.445(59.84)***

sqrt(BUSSEG)

+0.188(9.58)***

0.193(9.93)***

FOREIG

N+

0.612(15.79)***

0.583(15.00)***

RECIN

V+

0.271(5.36)***

0.305(6.02)***

ROA

��0

.330(�

8.67)***

�0.267(�

7.02)***

(Thetable

iscontinued

onthenextpage.)



TABLE

4(C

ontinued)

Dependentvariable

Sign

ln(AUDIT

FEE)

LOSS

+0.146(9.42)***

0.140(9.05)***

LEVERAGE

+�0

.219(�

5.25)

�0.198(�

4.77)

QUIC

K�

�0.026(�

5.39)***

�0.030(�

6.27)***

MW

+0.341(15.92)***

0.339(16.03)***

DED

IO+

0.152(1.55)*

0.153(1.57)*

BIG

N+

0.289(11.76)***

0.289(12.01)***

SPECIA

LIST

+�0

.001(�

0.06)

�0.003(�

0.19)

AUD

TENURE

?�0

.001(�

0.74)

�0.001(�

0.68)

YE

+0.086(4.62)***

0.084(4.54)***

AUD

OPIN

ION

+0.103(8.33)***

0.101(8.16)***

ln(NON

AUDIT

)+

0.091(14.12)***

0.089(13.82)***

Industry

&yearfixed

effects

Yes

Yes

Observations

11,837

11,837

Adjusted

R2

0.86

0.86

Notes:

Coeffi

cientvalues

(t-statistics)

are

shownwithstandard

errors

clustered

atthecompanylevel.**

*,**,*

indicate

statisticalsignificance

atthe1percent,5

percent,and10percentlevels,respectively,tw

o-tailed

(one-tailed

ifpredicted).Industry

fixed

effects

use

separate

indicators

forthe17industry

groupingsdefined

inFamaandFrench

1988.Refer

toTable

1forvariable

definitions.



AAER (p < 0.05), lawsuit (p < 0.05), and restatement (p < 0.01) models, respectively. Thepositive coefficient estimate for STRATEGY indicates that prospectors are more likely toexperience financial reporting irregularities than defenders. Results indicate that, ceterisparibus, the odds of experiencing an AAER, lawsuit, or restatement are restatement are2.32, 1.10, and 0.47 times higher, respectively, for firms with a STRATEGY score at thecutoff for prospectors relative to firms with a STRATEGY score at the cutoff for defend-ers. We note that although these irregularities are low-probability events, they are high-cost occurrences for the firms. The control variables across the samples are generally con-sistent with expectations from prior research. Overall, the model has a good fit with aROC curve value of 0.73 or greater for all models. Untabulated results suggest thatSTRATEGY significantly contributes to all three of the models’ explanatory power usinga likelihood ratio chi-squared test (p < 0.01) and using model fit statistics based on Raf-tery (1995).

Audit fee model

Table 4, panel B presents the results for our audit fee model. The positive and significantcoefficient for STRATEGY (p < 0.01) suggests that prospectors are associated with ahigher level of audit effort than defenders, consistent with Hypothesis 2. We note that thepositive and significant coefficient on STRATEGY suggests that STRATEGY captures ele-ments of client business risk that go beyond the risks captured by commonly used proxiesfor size, complexity, and risk (and other audit fee model controls). All the control variablecoefficients are significant and have the expected sign and significance except for SPE-CIALIST and AUD TENURE, which are insignificant.21 The sign on the coefficient forLEVERAGE is opposite our expectation. An economic interpretation of the coefficientestimate for STRATEGY indicates that, ceteris paribus, firms at the cutoff for prospectorswill pay audit fees approximately 22 percent higher than firms at the cutoff for defenders.The average audit fees in our sample are $1.62 million, suggesting the difference in auditfees paid by otherwise equal firms at the cutoff points for prospectors and defenders isapproximately $356,400. Therefore, these results suggest that auditors exert more auditeffort for companies on the prospector end of the strategy continuum.22 As an alternativemethod of quantifying the effect of STRATEGY, we compare predicted audit fees forprospector clients with and without STRATEGY in the fee model. We find that predictedaudit fees for prospector clients are, on average, $149,000 higher relative to all firms inour sample. Thus, our results suggest business strategy is an economically important deter-minant of audit effort.

7. Additional analyses

Reconciling the financial reporting irregularity and audit fee results

Our results suggest that even though auditors appear to exert greater audit effort for firmsemploying the prospector strategy, prospectors experience more financial reporting irregu-larities than defenders. One potential explanation for this apparent paradox is the timeperiod we examine. We examine financial reporting irregularities beginning in 1993; how-ever, our analysis of audit fees is restricted to periods after 2000. Thus, one explanationmight be a differential frequency of financial reporting irregularities for prospectors in the

21. To further investigate potential audit quality differences among prospector and defender clients, we also

examine whether industry specialist serves as a moderator on the effect of STRATEGY on audit fees by

including an interaction term between STRATEGY and SPECIALIST; the coefficient on the interaction is

not significant.

22. Untabulated results suggest that STRATEGY significantly contributes to the model’s explanatory power

(F-statistic 58.26, p < 0.001).



post-2000 period versus the 1990s and the sensitivity of audit fees to those financial report-ing irregularities.

Prior research demonstrates an increased sensitivity of audit fees to financial reportingrisk during 2000–2003. Specifically, Charles, Glover, and Sharp (2010) find that for theperiod 2000–2003 there was a significant increase in the association between audit fees andfinancial reporting risk, which they attribute to events that impacted the auditing profes-sion during that period (e.g., recommendations from the Blue Ribbon Committee onImproving the Effectiveness of Corporate Audit Committees, Enron, Worldcom, etc., Sar-banes-Oxley, and SAS No. 99). Thus, to explore the possibility that the sensitivity of auditfees to risk may differ before and after 2000, we align our financial reporting irregularitiesdata with our audit fee data by restricting financial reporting irregularities to post-2000.We find that our STRATEGY measure remains positive and significant in all our irregu-larity models in the post-2000 sample. In addition, even after adding controls for materialweakness occurrences and fees for nonaudit services in the financial reporting irregularitymodels, which restricts the sample to the post-2001 period, STRATEGY remains positiveand significant in all irregularity models.

We consider another potential explanation for our finding that prospector firmsappear to receive greater audit effort (as observed with higher audit fees) but experiencehigher incidences of financial reporting irregularities. Prior research has argued that theexpected positive relationship between client business risk and audit fees could arise fromthe implementation of more costly audit procedures to reduce audit risk (i.e., greater auditeffort) or from an assessed audit fee premium to cover the increase in auditor’s businessrisk (e.g., due to potential litigation). We try to disentangle these two explanations.23

If auditors are only assessing an audit fee premium to cover the increase in auditorbusiness risk and are not implementing the more costly audit procedures necessary tocover the additional risks of prospector clients, then prospectors could persist in experienc-ing irregularities despite the increase in audit fees. Thus, we consider whether the positiverelation between STRATEGY and financial reporting irregularities is moderated by loweraudit quality.

We create three additional indicator variables for low audit quality and interact eachwith STRATEGY in the financial irregularity model. The three indicator variables relateto different attributes of lower audit quality: low audit effort, low experience with theaudit client, and low auditor industry expertise.24 We measure low audit effort by con-structing an indicator variable equal to one if abnormal audit fees are negative (i.e., whereexpected fees exceed actual fees), following Choi, Kim, and Zang 2009 in the constructionof the abnormal fee measure. We measure low experience with the audit client by con-structing an indicator variable equal to one if audit firm tenure is below our samplemean.25 We measure low auditor industry expertise as an indicator variable equal to oneif the auditor is not an industry specialist. Untabulated results reveal that althoughSTRATEGY remains positive and significant, none of the interactions is significant, sug-gesting that the association between STRATEGY and financial reporting irregularities isnot moderated by lower audit quality.

We also consider whether auditors adjust their audit procedures for prospector clientswith irregularities relative to those without irregularities to determine if auditors do recog-nize the additional risks of clients that experience irregularities. To examine this issue, weconstruct a measure of abnormal audit fees where abnormal audit fees are defined as the

23. Refer to Stanley 2011 for a more in-depth discussion of these competing explanations.

24. We thank an anonymous reviewer for this suggestion.

25. We recognize that there is strong debate concerning whether audit tenure improves audit quality; however,

we take the position consistent with Myers, Myers, and Omer 2003 that greater tenure improves audit

quality.



difference between expected and actual audit fees (Choi et al. 2009). For our measure ofexpected fees we use the predicted values from our audit fee model that includes theSTRATEGY measure.26 We find that prospectors’ abnormal audit fees are negative over-all, suggesting that the assessed audit fees are too low on average. Furthermore, we findthat prospectors with and without irregularities do not incur statistically different abnor-mal audit fees. Thus, although auditors assess greater audit fees for prospector clients, theaudit fees do not appear sufficient to compensate for the additional risk of prospector cli-ents, including those with irregularities. Altogether, these findings suggest that audit feesmay be too constrained for prospector clients, especially those prospectors at a greater riskof experiencing irregularities.

Alternate measures of financial reporting irregularities

Instead of considering each financial reporting irregularity sample separately, Table 5 pre-sents results from combining our financial reporting irregularities such that our dependentvariable equals one if the firm experiences an AAER, lawsuit, and/or restatement and zerootherwise. Again consistent with Hypothesis 1, STRATEGY is positive and significant inthis model (p < 0.01). Results indicate that, ceteris paribus, the odds of experiencing anyirregularity are 0.62 times higher for firms with a STRATEGY score at the cutoff for pros-pectors relative to firms with a STRATEGY score at the cutoff for defenders.

As additional untabulated sensitivity tests, we first expand our investigation of irregu-larities to include a broader set of accounting-related SEC enforcement actions from theKarpoff, Lee, and Martin (2008a, 2008b) studies. We find that STRATEGY is positivelyassociated with these SEC enforcement actions (p < 0.05). We also investigate the associa-tion between STRATEGY and two commercial fraud risk measures (Accounting and Gov-ernance Risk and Accounting Risk) developed by Audit Integrity (see Price et al. 2011).We find that STRATEGY is positively related to both commercial risk measures(p < 0.01).

Alternate business strategy indicators

Our results are also robust in all the irregularity samples when we replace our discreteSTRATEGY measure with indicator variables for prospectors and defenders. The Prospec-tor indicator variable is positive and significant in the AAER, lawsuit, and restatementmodels. The coefficient for the Defender indicator is generally not significant in the irregu-larity models (except in the restatement model where the coefficient is negative and signifi-cant) suggesting that defenders are generally not significantly different from analyzers withrespect to the likelihood of experiencing irregularities.

Material weaknesses and nonaudit fees

We also include a control for material weakness occurrences in our irregularity modelsbecause material weaknesses have been linked to lower-quality accounting (e.g., Doyle,Ge, and McVay 2007a, 2007b). In addition, we control for nonaudit fees because anec-dotal evidence and corresponding legislation (e.g., Sarbanes-Oxley Act of 2002) suggestthat nonaudit services may impair audit quality.27 While imposing these controls restricts

26. For ease in interpreting the abnormal audit fee measure, we follow Choi et al. 2009 and “compute the dol-

lar value of abnormal fees as the differences between the actual dollar values of audit fees and normal dol-

lar values of audit fees after converting the estimated logged normal fees into their respective dollar values

(by using the exponential function to convert logged values to actual values)” (10). We then scale the dol-

lar value of abnormal fees by actual audit fees to obtain a percentage measure of abnormal fees.

27. We note that academic research has been mixed on this topic, and Kinney, Palmrose, and Scholz (2004)

and Prawitt, Sharp, and Wood (2012) do not find evidence to support the assertion that nonaudit services

impair audit quality.



our sample to fiscal year 2002 and onward, we note that although the coefficient on mate-rial weakness occurrences is positive and significant and the coefficient on nonaudit fees isinsignificant, the STRATEGY coefficient is positive and significant across all types of irreg-ularities.

Client business risk versus financial reporting risk

Given that our results suggest STRATEGY is associated with more frequent occurrencesof financial reporting irregularities, it is possible that STRATEGY proxies for companies’client business risk or financial reporting risk. To examine this issue, we conduct two tests.First, we estimate the predicted probabilities of each financial reporting irregularity with-out our STRATEGY measure in the models.28 We then add those predicted probabilitiesindividually in our audit fee model along with STRATEGY. This provides a more directtest of STRATEGY against financial reporting risk factors that are known to increaseaudit fees and allows us to determine whether STRATEGY captures client business risk orwhether STRATEGY is just a substitute for financial reporting risk. In untabulatedresults, we find that the coefficient for STRATEGY continues to be positive and significantin the audit fee model when including the predicted probabilities of financial reportingirregularities as additional controls. Second, we regress STRATEGY on the likelihood ofthe three financial reporting irregularities and use the residual from that estimation in ouraudit fee model in place of STRATEGY. This method should orthogonalize our STRAT-EGY measure from the likelihood of a financial reporting irregularity. We find the coeffi-cient on the residual from this estimation is positive and significant in the audit fee model,suggesting STRATEGY is not merely a substitute for financial reporting risk. Based onthese two tests, we suggest that our STRATEGY variable primarily represents client busi-ness risk.29

Firm size

Although our descriptive statistics for the audit fee model indicate no significant differ-ences between prospectors and defenders in either mean or median values of firm size (ln(ASSETS)), we examine whether firm size is a determinant of strategy and hence an alter-native explanation for our results.30 We use an approach based on Carson and Fargher2007 and form size quintiles based on ln(ASSETS) and interact STRATEGY with threeof the size quintiles representing the smallest, median, and largest companies in the sam-ple. We then include the following interaction terms in the audit fee model along with theSTRATEGY main effect: STRATEGY*SizeQ1, STRATEGY*SizeQ3, STRATEGY*SizeQ5(i.e., the interaction terms of STRATEGY with size quintiles 1, 3, and 5, respectively). Ourresults indicate that the coefficient on STRATEGY remains positive and significant in themodel (p < 0.01) while the coefficients on the interactions are as follows: STRAT-EGY*SizeQ1 is negative and significant, STRATEGY*SizeQ3 is insignificant, and STRAT-

28. Given the ROC values on estimates of the irregularity models without STRATEGY, we believe these are

reasonable in-sample estimates of the likelihoods of each of these outcomes.

29. We note that our audit fee model results are robust to controlling for bankruptcy risk using Altman’s Z-

score (Altman 1968), which suggests that STRATEGY does not substitute for bankruptcy risk. Specifically,

we find that the coefficient on Altman’s Z-score is negative and significant (p < 0.01) and the STRATEGY

coefficient remains positive and significant (p < 0.01) in our fee model.

30. The insignificant differences in size between defenders and prospectors in our audit fee sample are consis-

tent with prior research. Smith, Guthrie, and Chen (1989) find that there are few differences in the size dis-

tributions of defenders and prospectors. Further, in a factor analysis, we find that firm size does not load

on a factor with our business strategy components.



EGY*SizeQ5 is positive and significant. The results suggest that size is related but is not aprimary determinant of the STRATEGY results reported in Table 4.31 Finally, we useseemingly unrelated estimation (SUEST) to test whether the STRATEGY coefficientsacross the size quintiles are equal and find that the coefficients across size quintiles are notstatistically different (chi-square 5.60, p = 0.23). Therefore, based on these additional tests,we conclude that our results in Table 4, panel B, are not driven by effects related to firmsize.

Business strategy component analysis

To assess whether STRATEGY represents a construct separate from the measures of com-plexity and risk included in traditional audit fee models, we perform several additionaltests. First, we perform canonical correlation and redundancy index analyses (Stewart andLove 1968) on the raw STRATEGY components and the control variables from the auditfee model. We find evidence that the components of STRATEGY and the audit fee controlvariables comprise two separate constructs with very little overlapping variance betweenthem. The combination of the first two squared canonical correlations explains 93 percent

TABLE 5

Sensitivity analysis involving combined financial reporting irregularities

Dependent variable Sign FR_IRREGULARITY

Intercept ? �5.702 (�15.01)***STRATEGY + 0.040 (3.47)***

ln(ASSETS) ? 0.118 (4.53)***ROA � �0.002 (�0.04)LOSS + 0.284 (3.53)***

BTM � 0.033 (1.58)SALES GROWTH ? �0.000 (�1.23)M&A + �0.027 (�0.34)LEVERAGE ? �0.021 (�0.21)

FINANCING + 0.040 (0.29)FIRM AGE � �0.022 (2.83)***HERF + 0.594 (1.98)**

LITIGIOUS + 0.196 (1.90)**DAP ? �0.372 (�1.70)*BIGN � �0.038 (�0.35)

DED IO ? 1.704 (3.90)***Industry & year fixed effects YesObservations 16,290McFadden’s (pseudo) R2 0.114

ROC curve 0.74

Notes:

Coefficient values (z-statistics) are shown with standard errors clustered at the company level. ***,**,* indicate statistical significance at the 1 percent, 5 percent, and 10 percent levels,

respectively, two-tailed (one-tailed if predicted). The dependent variable is an indicator variable

equal to one if the company misreported its financial results during the current year according

to an AAER, lawsuit, and/or restatement, and zero otherwise. Industry fixed effects use

separate indicators for the 17 industry groupings defined in Fama and French 1988. Refer to

Table 1 for variable definitions.

31. We also estimate the audit fee model in each size quintile and the coefficient on STRATEGY remains posi-

tive and significant in all quintiles.



of the covariance between the two variable sets (i.e., STRATEGY components and theaudit fee control components); however, an average of only 7 percent of each variable set’svariance is included in these correlations (i.e., 8.8 percent of the variance in the STRAT-EGY components are explained by the audit control variables while only 5.5 percent ofthe variance in the audit fee controls are explained by the STRATEGY components).

Next, we deconstruct the STRATEGY measure into its individual components andreplace STRATEGY with its individual components in the audit fee and financial report-ing irregularity models. We find that the coefficients on many of the individual compo-nents are not significant or are significant but the components’ signs are not alwaysconsistent with the sign of the composite STRATEGY measure.32 In addition, when weperform factor analysis on all six raw components of STRATEGY, all items load on onefactor. To be consistent with organizational theory which evaluates these components rela-tive to industry competitors, we create factor scores by industry-year. In untabulated tests,we replace our discrete STRATEGY measure in our regression models with factor scorescreated from the six STRATEGY components and our inferences remain unchanged. Thisprovides confirming evidence that when we use loadings based on factor analysis, the con-struct we capture with our STRATEGY measure is confirmed. Collectively, our compo-nent analyses suggest that our STRATEGY measure captures a construct that is “greaterthan the sum of its parts”.

Finally, we perform a factor analysis on our STRATEGY composite score with ouraudit fee control variables and find that the audit fee controls load primarily on the firsttwo factors while STRATEGY loads weakly on the third factor (factor loading < 0.40),separate from all audit fee controls except for FOREIGN, RECINV, and QUICK.33 Thus,STRATEGY appears to capture a construct separate from individual measures of clientsize, complexity, and risk and is not a substitute for these measures.

Company strategy consistency

Organizational theory (Miles and Snow 1978, 2003) posits that when companies adopt a par-ticular business strategy, the strategy should remain consistent over time. Thus, we expectprospectors, defenders, and analyzers to have relatively stable STRATEGY scores from yearto year. For example, we expect companies classified as prospectors to demonstrate highSTRATEGY scores and companies classified as defenders to exhibit low STRATEGY scoresconsistently during our sample period. To examine the consistency of our STRATEGYscores, we first analyze the variance of our STRATEGY measure within companies, whichrequires that companies have at least two consecutive years to compute variance.

Untabulated results indicate that about 5 percent of companies have STRATEGYscores that never change. The mean (median) variance is 2.87 (1.75). We run first differ-ences for each company-year STRATEGY score to determine how many total times com-panies with consecutive year observations changed their scores. About 34 percent ofcompanies did not change scores from year to year, while about 42 percent of companieschanged their scores by only one value (e.g., changing from 30 to 29). Less than 3 percentof companies changed their STRATEGY score by more than three values. Untabulated

32. For example, in the audit fee model, employee fluctuations are positive and significant (r(EMP5)), while

capital intensity (CAP5) is negative and significant; none of the other variables is significant. In the AAER

model, the coefficients on both SG&A ratio (SGA5) and capital intensity (CAP5) are negative and signifi-

cant; none of the variables is positively related to AAERs. Similar results are found for the other two

irregularity models.

33. Controls related to client size and auditor quality load on a separate factor from STRATEGY (e.g., ln

(ASSETS), sqrt(BUS SEG), DED IO, BIGN, SPECIALIST, AUD TENURE, AUD OPINION, and ln

(NON AUDIT)). The other remaining audit fee controls also load on the second factor separate from

STRATEGY.



TABLE 6

Sensitivity analysis including investment opportunity sets (IOS)

Panel A: Financial reporting irregularities models

Dependentvariable Sign AAER LAWSUIT RESTATEMENT

Intercept ? �11.026 (�6.83)*** �8.833 (�9.27)*** �5.797 (�13.75)***

STRATEGY + 0.091 (1.68)** 0.066 (2.27)** 0.038 (3.02)***IOS ? 0.846 (0.43) 1.382 (1.10) �1.825 (�2.77)***ln(ASSETS) ? 0.315 (3.76)*** 0.368 (7.46)*** 0.039 (1.36)ROA � �0.168 (�3.49)*** �0.152 (�2.67)*** 0.099 (1.30)

LOSS + 0.300 (1.12) 0.180 (1.08) 0.267 (2.99)***BTM � 0.083 (0.92) 0.018 (0.46) 0.038 (1.60)SALES GROWTH ? 0.000 (0.39) 0.000 (0.52) �0.000 (�1.23)

M&A + 0.601 (2.15)** 0.221 (1.17) �0.040 (�0.45)LEVERAGE ? �0.796 (�1.30) �0.433 (�1.19) 0.063 (0.71)FINANCING + �0.500 (�0.82) 0.436 (1.50)* 0.012 (0.08)

FIRM AGE � 0.003 (0.15) �0.070 (�3.47)*** �0.013 (�1.59)*HERF + 2.224 (2.65)*** 2.548 (3.38)*** 0.337 (0.80)LITIGIOUS + 0.153 (0.41) 0.713 (2.69)*** 0.367 (2.67)***DAP ? 0.153 (0.16) �0.395 (�0.86) �0.515 (�2.15)**

BIGN � �0.131 (�0.32) �0.867 (�3.84)*** 0.163 (1.37)DED IO ? 0.718 (0.50) 2.031 (2.28)** 1.608 (3.31)***Industry &

year fixed effects

Yes Yes Yes

Observations 11,322 14,915 13,847McFadden’s

(pseudo) R20.115 0.127 0.104

ROC curve 0.79 0.80 0.73

Notes:

Coefficient values (z-statistics) are shown with standard errors clustered at the company level. ***,**,* indicate statistical significance at the 1 percent, 5 percent, and 10 percent levels,

respectively, two-tailed (one-tailed if predicted). The dependent variable is an indicator variable

equal to one if the company misreported its financial results during the current year according

to an AAER, lawsuit, or restatement, and zero otherwise. IOS refers to investment

opportunity set following Cahan et al. 2008. Industry fixed effects use separate indicators for

the 17 industry groupings defined in Fama and French 1988. Refer to Table 1 for variable

definitions.

Panel B: Audit fee model

Dependent variable Sign ln(AUDIT FEE)

Intercept ? 7.588 (55.59)***STRATEGY + 0.016 (6.66)***IOS + 0.491 (4.85)***

ln(ASSETS) + 0.448 (58.94)***sqrt(BUS SEG) + 0.196 (9.76)***FOREIGN + 0.547 (13.55)***




correlation tests reveal that STRATEGY is positive and significantly correlated with itsone-year lag value and has a correlation coefficient above 0.90 (p < 0.001) for both Pear-son and Spearman correlations. Altogether, consistent with expectations from theory,these results suggest that our STRATEGY measure is stable over time within firms.

Investment opportunity sets

In a final additional analysis, we control for investment opportunity sets (IOS) becauseprior research has found that audit clients with higher IOS are more likely to have higher-quality industry-specialist auditors and pay higher audit fees (Cahan, Godfrey, Hamilton,and Jeter 2008). Investment opportunity sets, as operationalized by Cahan et al. 2008 andbased on the IOS measure developed by Baber, Janakiraman, and Kang 1996, captureboth industry-level and firm-level IOS; thus, it is likely that STRATEGY, if it is a substi-tute, substitutes for firm-level IOS. We estimate the IOS for our sample firms and includethe measure in our audit fee and financial reporting irregularity models. We note that bothprospector and defender strategies exist at various levels of our IOS measure, includingindustries where the IOS is the highest. As shown in Table 6, panels A and B, the coeffi-cient on STRATEGY remains positive and significant after controlling for IOS in ourirregularity and audit fee models. Thus, our results suggest that STRATEGY is not aproxy for investment opportunities.

8. Conclusion

Using Miles and Snow’s (1978, 2003) strategy typology, we provide a measure of businessstrategy that requires only publicly available information and is generalizable across indus-tries. Using this measure, we first investigate whether companies’ business strategies exhibitdifferences in the occurrence of financial reporting irregularities. We find that companies fol-

TABLE 6 (Continued)

Dependent variable Sign ln(AUDIT FEE)

RECINV + 0.308 (6.04)***

ROA � �0.253 (�6.49)***LOSS + 0.138 (8.72)***LEVERAGE + �0.187 (�4.38)QUICK � �0.036 (�7.38)***

MW + 0.343 (15.94)***DED IO + 0.178 (1.83)**BIGN + 0.297 (12.09)***

SPECIALIST + 0.005 (0.28)AUD TENURE ? �0.001 (�0.63)YE + 0.086 (4.57)***

AUD OPINION + 0.100 (7.93)***ln(NON AUDIT) + 0.085 (13.06)***Industry & year fixed effects YesObservations 11,147

Adjusted R2 0.86

Notes:

Coefficient values (t-statistics) are shown with standard errors clustered at the company level. ***,**,* indicate statistical significance at the 1 percent, 5 percent, and 10 percent levels, respectively,

two-tailed (one-tailed if predicted). IOS refers to investment opportunity set following Cahan

et al. 2008. Industry fixed effects use separate indicators for the 17 industry groupings defined

in Fama and French 1988. Refer to Table 1 for variable definitions.



lowing a prospector strategy are more likely than companies following a defender strategy toexperience financial reporting irregularities across three samples of irregularities: SEC AA-ERs, shareholder lawsuits related to alleged accounting improprieties, and accountingrestatements. Based on additional analysis, we provide evidence that the business strategymeasure represents client business risk and is not a substitute for financial reporting risk.

Next, we explore whether clients’ business strategies represent an underlying determi-nant of audit effort levels because of differences in the client business risk of the strategies.We find that clients following a prospector strategy have higher audit fees, suggesting thatauditors expend greater audit effort for these clients compared to defenders. Even aftercontrolling for standard measures of client size, risk, and complexity, we find that ourbusiness strategy measure is significant in an audit fee model. In additional analyses weprovide evidence that business strategy provides incremental explanatory power over mea-sures suggested by prior literature as determinants of audit fees.

We consider several alternative explanations for why prospectors experience irregulari-ties despite the apparent increase in auditor effort and conclude that lower audit qualitydoes not explain our results. Instead, our findings suggest despite the higher audit fees(and higher audit effort) for prospectors, fees are not high enough to account for the riski-ness of these clients. Alternatively, there may be nothing auditors can do to further reducethis risk with current audit technology. Future research should explore how to improveaudits of prospector clients.

Finally, we deconstruct the business strategy measure into its individual componentsin our audit fee and financial reporting irregularities models and find that the individualmeasures either do not load significantly or load with opposite signs. In addition, wereplace our STRATEGY measure with factor loadings and obtain similar results. Thus, itappears that using organizational strategy typology to combine these measures captures aunique construct that is “greater than the sum of its parts”.

Our research is subject to several limitations. Although we rely on Miles and Snow’s(1978, 2003) strategy typology to create our STRATEGY measure, we assess businessstrategy with noise. To the extent that measurement error could lead to misclassifyingsome firms’ business strategies, this is a limitation of our study. Another limitation is ourinability to determine precisely why prospector firms experience more financial reportingirregularities in spite of greater audit effort. While we explored several potential explana-tions for these apparently paradoxical findings, additional research is needed to investigatethese findings further. Finally, as noted previously, our results regarding financial report-ing irregularities should be interpreted carefully as these irregularities (e.g., AAERs,accounting-related lawsuits, and restatements) are relatively infrequent events.

We make several contributions to the existing accounting literatures on financialreporting irregularities and audit effort. First, we provide evidence that differences in theclient business risk of companies’ choice of business strategies is an underlying determi-nant of the likelihood of financial reporting irregularities and thus is a determinant offinancial statement quality. Second, our results based on an audit fee model provide evi-dence that auditors appear to recognize and adjust audit effort based on client businessstrategies, which is a broader, more comprehensive business risk approach in assessingtheir audit clients. Business strategy represents features of client business risk beyond whathas been represented in the literature using traditional individual proxies for client size,risk, and complexity. Finally, we construct a comprehensive, theory-based business strat-egy measure that is replicable using publicly available data. This is important becauseprior measures have required access to management and/or proprietary data. Our studyprovides evidence that is potentially useful to auditors, regulators, investors, and analystsbecause it identifies organizational business strategies as an important determinant of bothfinancial reporting irregularities and audit effort.



Appendix 1

Business strategy characteristics

Prospector Defender

Definition Company that continually

seeks new and innovativeproducts and operates onthe basis of a diversified

decision maker model.

Company that is typically

vertically integrated, hasa narrow set of decisionmarkers, specializes in

a very narrow product lineand focuses heavily oncost reduction.

Competitive advantage Market innovation. Efficiency and stability.Competitive disadvantage Risk of low profitability

and overextension of

resources.

Adaptability to market shiftand threat of obsolescence.

Research and development Extensive R&D in order to

exploit new product andmarket opportunities.

Minimal R&D and is

usually closely related tocurrent products.

Efficiency Never achieve maximumefficiency in production anddistribution systems.

Achieve efficiency inproduction and distributionsystems.

Growth Growth occurs in spurtsthrough product-market

development.

Cautious and incrementalgrowth through market

penetration.

Marketing Strong focus on marketingfunction.

Weak focus on marketingfunction while emphasis ison production and financialfunctions.

Organizational structure and stability Decentralized control tofacilitate and coordinate

diverse/numerous operations.Focus on product groups.Dominant coalition is

transitory and may hirefrom outside.

Strict centralized control toensure efficiency and

focuses on functionaldivisions. The dominantcoalition is lengthy and

tends to promote fromwithin.

Capital intensity Low degree of mechanizationand routinization to avoidlengthy commitments to

single technological process.

High degree ofmechanization androutinization focusing on

single core cost-efficienttechnology.



Appendix 2

Business strategy composite measure construction

The STRATEGY measure is constructed of the following six measures based on Ittner et al. 1997

and Miles and Snow 1978 and 2003. Each of the variables is measured per company-year based onthe rolling prior five-year average. Then each of these average variables is ranked into quintiles perindustry (two-digit SIC code) and year. Those observations in the highest quintiles are given a score

of 5, those in the second highest quintile are given a score of 4, and so on, while those observationsin the lowest quintiles are given a score of 1 (except capital intensity which is reversed-scored so thatobservations in the lowest (highest) quintile are given a score of 5 (1)). The scores are summed over

the six measures per company-year such that a company could have receive a maximum score of 30(prospector-type) and a minimum score of 6 (defender-type). Therefore, our discrete STRATEGYscore ranges along a continuum in value from 6 to 30 with defender- and prospector-type companiescloser to the endpoints and analyzer-type companies constituting the middle of the STRATEGY

continuum, consistent with organizational theory (Miles and Snow 1978, 2003). Although ourdiscrete STRATEGY measure is our primary measure, we also consider the following strictdefinitions of STRATEGY-types: defenders (6–12); analyzers (13–23); prospectors (24–30).

Variable measures

Variable measurement

[COMPUSTAT code]

(1) Ratio of research and development

to sales (RDS5)Company’s propensity to search for new products.

Ratio of research and development expenditures

[XRD] to sales [SALE] computed over a rollingprior five-year average.

(2) Ratio of employees to sales (EMPS5)Company’s ability to produce and distributeproducts and services efficiently.

Ratio of the number of employees [EMP] tosales [SALE] computed over a rolling priorfive-year average.

(3) Change in total revenue (REV5)Company’s historical growth or investment

opportunities.

One-year percentage change in total sales[SALE] computed over a rolling prior five-year

average.

(4) Marketing to sales (SGA5)

Company’s focus on exploiting new productsand services.

Ratio of selling, general and administrative

expenses [XSGA] to sales [SALE] computedover a rolling prior five-year average.

(5) Employee fluctuations (r(EMP5))Company’s organizational stability.

Standard deviation of the total number ofemployees [EMP] computed over a rollingprior five-year period.

(6) Capital intensity (CAP5)Company’s commitment to technological

efficiency.

Capital intensity which is measured as netPPE [PPENT] scaled by total assets [AT]

computed over a rolling prior five-year average.



Appendix 3

Business strategy type examples

Miles and Snow (1994) provide descriptive examples of companies illustrating Miles-Snow strategictypes and we compare some of these examples to our own strategy classification (all quotes in the

Table below are from Miles and Snow 1994 (henceforth M&S), unless otherwise indicated). We alsocompare our strategy classification to strategic information from corporate websites. In oursensitivity analysis, we replace our discrete STRATEGY measure with indicator variables as follows:

defenders are defined with a STRATEGY range of 6–12; analyzers are defined with a STRATEGYrange of 13–23; and prospectors are defined with a STRATEGY range of 24–30.

Strategy type RationaleOur classification(STRATEGY)

Prospector:

“Some firms achieve successby being first, either byanticipating where themarket is going or by

shaping the market’sdirection through their ownresearch and development

efforts. We call these firms‘Prospectors’ because theycontinually search for new

products, services,technologies, and markets”(M&S, 12).

“Cisco is differentiated fromits peers and has a uniquestrategy to grow faster thanthe market. Our long-held

leadership position in routingand switching is well known.We work to protect and

extend this leadership yearafter year by investing morein ongoing innovation than

do any of our peers. This is acore focus of our business.Our ability to innovate iswhy customers rely on us as

a strategic business partnersrather than merely a producttechnology partner.” (Cisco

Systems 2010)

Cisco Systems is classified as aprospector and upper-rangeanalyzer during the periodanalyzed (1995–2009).




Appendix 3 (Continued)


“Successful firms not onlydevelop clear decision-making and performancecriteria, they usually build

into these a developmentalcomponent” (M&S, 178).

“3M is fundamentally ascience-based company. Weproduce thousands ofimaginative products, and

we’re a leader in scores ofmarkets — from health careand highway safety to office

products and abrasive andadhesives” (3M Company2011). “For first-to-market

Prospector firms like 3M

Company... myths aboutdistinctive competenceusually turn on displays of

almost rebellious creativity”(M&S, 172). “3M Company

has for some time built a

measure of productinnovation into itsperformance review

procedure. A minimum of 25percent of the annual sales ofevery division is expected tobe generated from products

introduced within the pastfive years. At 3M, that policyleaves no question about

whether new ideas are to bestimulated and supported —they must be to meet division

performance standards”(M&S, 178).

3M Company is classified as aprospector and upper-rangeanalyzer (i.e., score of 23 justoutside of prospector range)

during period analyzed (1993–2009).

Defender:“Defenders usually do not

attempt to operate across awide product or servicearena. Instead, they search

for economies of scale inthose areas that are relativelyhealthy, stable, and

predictable” (M&S, 12).

“Founded in 1902, Lamar

Advertising operates over 150outdoor advertisingcompanies and 63 companies.

They reach driving audiencesacross the United States,Canada and Puerto Rico

through billboards, digitalbillboards, bus shelters,benches and buses. Lamar isalso the nation’s leader in the

highway logo sign business”(Lamar Advertising 2011).

Lamar Advertising Company is

classified as a defender andlow-range analyzer duringthe period analyzed (2001–2009).




Appendix 3 (Continued)


“Defenders are innovative indelivering an existing line ofproducts and services to theircustomers” (M&S, 13).

“American Pacific Corporation

was founded … for thepurpose of manufacturingspecialty chemicals,

principally oxidizers. Sincethen, our business hasexpanded to focus on three

primary business segments:Specialty Chemicals, FineChemicals and Aerospace

Equipment... AMPAC isrecognized as a leadingmanufacturer of specialtyand fine chemicals within our

focused markets.” “Throughtechnical and manufacturingexpertise and a focus on

customer service, AMPAChas achieved a reputation forquality, reliability, technical

performance and innovation”(American PacificCorporation 2010).

American Pacific Corporation

is classified as a defenderexcept in the recent yearswhen becoming a lower-

range analyzer during theperiod analyzed (1993–2009).

References

3M Company. 2011. Company information: Who we are. Available online at http://solutions.3m.com/

wps/portal/3M/en_US/about-3M/information/about/us, retrieved August 6, 2011.

Ali, A., and S. Kallapur. 2001. Securities price consequences of the Private Securities Litigation

Reform Act of 1995 and related events. The Accounting Review 76 (3): 431–60.

Altman, E. 1968. Financial ratios, discriminant analysis and the prediction of corporate bankruptcy.

The Journal of Finance 23 (4): 589–609.

American Institute of Certified Public Accountants (AICPA). 2002. Statement on Auditing Standards

(SAS) No. 99: Consideration of fraud in a financial statement audit. New York: AICPA.

American Institute of Certified Public Accountants (AICPA). 2006. AU Section 311: Planning and

supervision. New York: AICPA.

American Pacific Corporation. 2010. About us. Available online at http://www.apfc.com/overview.

php, retrieved August 8, 2011.

Baber, W. R., S. N. Janakiraman, and S.-H. Kang. 1996. Investment opportunities and the structure

of executive compensation. Journal of Accounting and Economics 21 (3): 297–318.

Beasley, M. S., J. V. Carcello, and D. R. Hermanson. 1999. Fraudulent financial reporting1987–1997:

An analysis of U.S. public companies. Committee of Sponsoring Organizations of the Treadway

Commission. Available at http://www.coso.org

Bell, T. B., R. Doogar, and I. Solomon. 2008. Audit labor usage and fees under business risk audit-

ing. Journal of Accounting Research 46 (4): 729–60.

Beneish, M. D. 1997. Detecting GAAP violation: Implications for assessing earnings management

among firms with extreme financial performance. Journal of Accounting and Public Policy 16 (3):

271–309.



Beneish, M. D. 1999. Incentives and penalties related to earnings overstatements that violate GAAP.

The Accounting Review 74 (4): 425–57.

Bentley, K. A. 2012. Antecedents to financial statement misreporting: The influence of organizational

business strategy, ethical culture, and climate. Working paper, Texas A&M University.

Burns, N., and S. Kedia. 2006. The impact of performance-based compensation on misreporting.

Journal of Financial Economics 79 (1): 35–67.

Bushee, B. J. 2001. Do institutional investors prefer near-term earnings over long-run value? Con-

temporary Accounting Research 18 (2): 207–46.

Bushee, B. J., and C. Noe. 2000. Corporate disclosure practices, institutional investors, and stock

return volatility. Journal of Accounting Research 38 (Supplement): 171–202.

Cahan, S. F., J. M. Godfrey, J. Hamilton, and D. C. Jeter. 2008. Auditor specialization, auditor

dominance, and audit fees: The role of investment opportunities. The Accounting Review 83 (6):

1393–423.

Carson, E., and N. Fargher. 2007. Note on audit fee premiums to client size and industry specializa-

tion. Accounting and Finance 47: 423–46.

Casterella, J. R., J. R. Francis, B. L. Lewis, and P. L. Walker. 2004. Auditor industry specialization,

client bargaining power, and audit pricing. Auditing: A Journal of Practice and Theory 23 (1):

123–40.

Charles, S. L., S. M. Glover, and N. Y. Sharp. 2010. The association between financial reporting

risk and audit fees before and after the historic events surrounding SOX. Auditing: A Journal of

Practice and Theory 29 (1): 15–39.

Cheng, Q. 2005. What determines residual income? The Accounting Review 80 (1): 85–112.

Choi, J.-H., J.-B. Kim, and Y. Zang. 2009. Do abnormally high audit fees impair audit quality?

Working paper. Available at SSRN.

Cisco Systems. 2010. Annual report. http://www.cisco.com/

Cobbin, P. E. 2002. International dimensions of the audit fee determinants literature. International

Journal of Auditing 6 (1): 53–77.

Collins, F., O. Holzmann, and R. Mendoza. 1997. Strategy, budgeting, and crisis in Latin America.

Accounting, Organizations and Society 22 (7): 669–89.

DeAngelo, L. E. 1981. Auditor size and audit quality. Journal of Accounting and Economics 3 (3):

183–99.

Dechow, P. M., and I. D. Dichev. 2002. The quality of accruals and earnings: The role of accrual

estimation errors. The Accounting Review 77 (Supplement): 35–59.

Dechow, P. M., W. Ge, C. R. Larson, and R. G. Sloan. 2011. Predicting material accounting mis-

statements. Contemporary Accounting Research 28 (1): 17–82.

Dechow, P. M., R. G. Sloan, and A. P. Sweeney. 1996. Causes and consequences of earnings manip-

ulation: An analysis of firms subject to enforcement actions by the SEC. Contemporary Account-

ing Research 13 (1): 1–36.

Dent, J. F. 1990. Strategy, organization and control: Some possibilities for accounting research.

Accounting, Organizations and Society 15 (1–2): 3–25.

Doogar, R., P. Sivadasan, and I. Solomon. 2010. The regulation of public company auditing: Evi-

dence from the transition to AS5. Journal of Accounting Research 48 (4): 795–814.

Doyle, J., W. Ge, and S. McVay. 2007a. Accruals quality and internal control over financial report-

ing. The Accounting Review 82 (5): 1141–70.

Doyle, J., W. Ge, and S. McVay. 2007b. Determinants of weaknesses in internal control over finan-

cial reporting. Journal of Accounting and Economics 44 (1–2): 193–223.

Efendi, J., A. Srivastava, and E. P. Swanson. 2007. Why do corporate managers misstate financial

statements? The role of option compensation and other factors. Journal of Financial Economics

85 (35): 667–708.

Erickson, M., M. Hanlon, and E. Maydew. 2006. Is there a link between executive equity incentives

and accounting fraud? Journal of Accounting Research 44 (1): 113–43.



Fama, E. F., and K. R. French. 1988. Permanent and temporary components of stock prices. The

Journal of Political Economy 96 (2): 246–73.

Farber, D. B. 2005. Restoring trust after fraud: Does corporate governance matter? The Accounting

Review 80 (2): 539–61.

Francis, J., D. Philbrick, and K. Schipper. 1994. Shareholder litigation and corporate disclosures.

Journal of Accounting Research 32 (2): 137–64.

Francis, J. R., K. Reichelt, and D. Wang. 2005. The pricing of national and city-specific reputations

for industry expertise in the U.S. audit market. The Accounting Review 80 (1): 113–36.

Hackenbrack, K., and W. R. Knechel. 1997. Resource allocation decisions in audit engagements.

Contemporary Accounting Research 114 (3): 481–49.

Hambrick, D. C. 1981. Environment, strategy, and power within top management teams. Adminis-

trative Science Quarterly 26 (2): 253–75.

Hambrick, D. C. 1983. Some tests of the effectiveness and functional attributes of Miles and Snow’s

strategic types. The Academy of Management Journal 26 (1): 5–26.

Hay, D. C., W. R. Knechel, and N. Wong. 2006. Audit fees: A meta-analysis of the effect of supply

and demand attributes. Contemporary Accounting Research 23 (1): 141–91.

Higgins, D., T. C. Omer, and J. D. Phillips. 2011. Is organizational strategy a determinant of tax

avoidance? Working paper, University of Connecticut.

Hogan, C. E., Z. Rezaee, R. A. Riley, and U. K. Velury. 2008. Financial statement fraud:

Insights from the academic literature. Auditing: A Journal of Practice and Theory 27 (2): 231–

52.

Hogan, C. E., and M. S. Wilkins. 2008. Evidence on the audit risk model: Do auditors increase

audit fees in the presence of internal control deficiencies? Contemporary Accounting Research 25

(1): 219–42.

Hribar, P., T. Kravet, and R. Wilson. 2009. A new measure of accounting quality. Working paper.

Available at SSRN.

International Federation of Accountants (IFAC). 2009. International Standard on Auditing (ISA)

315: Identifying and assessing the risks of material misstatement through understanding the entity

and its environment. New York: IFAC. Available at http://web.ifac.org/clarity-center/isa-315.

Ittner, C. D., D. F. Larcker, and M. V. Rajan. 1997. The choice of performance measures in annual

bonus contracts. The Accounting Review 72 (2): 231–55.

Jan, C.-L., and J. A. Ou. 2008. Negative book value firms and their valuation. Working paper.

Available at SSRN.

Johnstone, K. M. 2000. Client-acceptance decisions: Simultaneous effects of client business risk,

audit risk, auditor business risk, and risk adaptation. Auditing: A Journal of Practice and Theory

19 (1): 1–25.

Johnstone, K. M., and J. C. Bedard. 2003. Risk management in client acceptance decisions. The

Accounting Review 78 (4): 1003–025.

Jones, K. L., G. V. Krishnan, and K. D. Melendrez. 2008. Do models of discretionary accruals

detect actual cases of fraudulent and restated earnings? An empirical analysis. Contemporary

Accounting Research 25 (2): 499–531.

Karpoff, J. M., D. S. Lee, and G. S. Martin. 2008a. The consequences to managers for financial mis-

representation. Journal of Financial Economics 88 (2): 193–215.

Karpoff, J. M., D. S. Lee, and G. S. Martin. 2008b. The cost to firms of cooking the books. Journal

of Financial and Quantitative Analysis 43 (3): 581–612.

Kinney, W. R., Z.-V. Palmrose, and S. Scholz. 2004. Auditor independence, non-audit services and

restatements: Was the U.S. government right? Journal of Accounting Research 42 (3): 561–88.

Knechel, W. R., P. Rouse, and C. Schelleman. 2009. A modified audit production framework: Eval-

uating the relative efficiency of audit engagements. The Accounting Review 84 (5): 1607–638.

Lamar Advertising. 2011. About us. Available online at http://www.lamar.com/About, retrieved

August 6, 2011.



Langfield-Smith, K. 1997. Management control systems and strategy: A critical review. Accounting,

Organizations and Society 22 (2): 207–32.

Larcker, D. F., and S. A. Richardson. 2004. Fees paid to audit firms, accrual choices, and corporate

governance. Journal of Accounting Research 42 (3): 625–58.

Loebbecke, J. K., M. M. Eining, and J. J. Willingham. 1989. Auditors’ experience with material

irregularities: Frequency, nature, and detectability. Auditing: A Journal of Practice and Theory 9

(1): 1–28.

March, J. G. 1991. Exploration and exploitation in organizational learning. Organization Science 2

(1): 71–87.

McGuire, S. T., T. C. Omer, and N. Y. Sharp. 2012. The impact of religion on financial reporting

irregularities. The Accounting Review 87 (2): 645–73.

Miles, R. E., and C. C. Snow. 1978. Organizational strategy, structure and process. New York:

McGraw-Hill.

Miles, R. E., and C. C. Snow. 1994. Fit, failure, and the hall of fame: How companies succeed or fail.

New York: The Free Press.

Miles, R. E., and C. C. Snow. 2003. Organizational strategy, structure, and process. Stanford, CA:

Stanford University Press.

Myers, J. N., L. A. Myers, and T. C. Omer. 2003. Exploring the term of the auditor-client relation-

ship and the quality of earnings: A case for mandatory auditor rotation? The Accounting Review

78 (3): 779–99.

O’Keefe, T. B., D. A. Simunic, and M. T. Stein. 1994. The production of audit services: Evidence

from a major public accounting firm. Journal of Accounting Research 32 (2): 241–61.

Palmrose, Z.-V. 1988. An analysis of auditor litigation and audit service quality. The Accounting

Review 63 (1): 55–73.

Petersen, M. A. 2009. Estimating standard errors in finance panel data sets: Comparing approaches.

The Review of Financial Studies 22 (1): 435–80.

Porter, M. E. 1980. Competitive strategy: Techniques for analyzing industries and competitors. New

York: The Free Press.

Prawitt, D. F., N. Y. Sharp, and D. A. Wood. 2012. Internal audit outsourcing and the risk of mis-

leading or fraudulent financial reporting: Did Sarbanes-Oxley get it wrong? Contemporary

Accounting Research, 29 (4): 1109–1136.

Price, R. A., N. Y. Sharp, and D. A. Wood. 2011. Detecting and predicting accounting irregularities:

A comparison of commercial and academic risk measures. Accounting Horizons 25 (4): 755–80.

Public Company Accounting Oversight Board (PCAOB). 2007. Auditing Standard No. 5: An audit of

internal control over financial reporting that is integrated with an audit of financial statements and

related independence rule and conforming amendments. Release No. 2007-005A, June 12. Wash-

ington, DC: PCAOB.

Raftery, A. E. 1995. Bayesian model selection in social research. Sociological Methodology 25: 111–

63.

Raghunandan, K., and D. V. Rama. 2006. SOX section 404 material weakness disclosures and audit

fees. Auditing: A Journal of Practice and Theory 25 (1): 99–114.

Rajagopalan, N. 1997. Strategic orientations, incentive plan adoptions, and firm performance: Evi-

dence from electric utility firms. Strategic Management Journal 18 (10): 761–85.

Reichelt, K. J., and D. Wang. 2010. National and office-specific measures of auditor industry exper-

tise and effects on audit quality. Journal of Accounting Research 48 (3): 647–86.

Sarbanes-Oxley Act of 2002. Pub. L. No. 107-204, 116 Stat. 745.

Schelleman, C., and W. R. Knechel. 2010. Short-term accruals and the pricing and production of

audit services. Auditing: A Journal of Practice and Theory 29 (1): 221–50.

Seifzadeh, P. 2011. Business strategy and the synergistic combination of exploration and exploita-

tion. Working paper, University of Western Ontario.



Simons, R. 1987. Accounting control systems and business strategy: An empirical analysis. Account-

ing, Organizations and Society 12 (4): 357–74.

Simunic, D. A. 1980. The pricing of audit services: Theory and evidence. Journal of Accounting

Research 18 (1): 161–90.

Singh, P., and N. C. Agrawal. 2002. The effects of firm strategy on the level and structure of execu-

tive compensation. Canadian Journal of Administrative Sciences 19 (1): 42–56.

Smith, K. G., J. P. Guthrie, and M.-J. Chen. 1989. Strategy, size and performance. Organization

Studies 10 (1): 63–81.

Stanley, J. D. 2011. Is the audit fee disclosure a leading indicator of clients’ business risk? Auditing:

A Journal of Practice and Theory 30 (3): 157–79.

Stewart, D., and W. Love. 1968. A general canonical correlation index. Psychological Bulletin 70 (3):

160–63.

Summers, S. L., and J. T. Sweeney. 1998. Fraudulently misstated financial statements and insider

trading: An empirical analysis. The Accounting Review 73 (1): 131–46.

Treacy, M., and F. Wiersema. 1995. The discipline of market leaders. Reading, MA: Addison-Wes-

ley.

Turner, L. E., and T. R. Weirich. 2006. A closer look at financial statement restatements: Analyzing

the reasons behind the trend. The CPA Journal, December.

Zahra, S. A., R. L. Priem, and A. A. Rasheed. 2005. The antecedents and consequences of top man-

agement fraud. Journal of Management 31 (6): 803–28.



business strategy, financial reporting irregularities, and audit...

Documents