Inexperienced Component Auditors and the Internal Information Asymmetry of Multinational Corporations

Abstract

With the global expansion of business operations, many U.S. corporations now operate internationally. While these multinational corporations (MNCs) rely on information about their operations to allocate resources, this information is likely dispersed within the firm, leading to internal information asymmetry between top executives and divisional managers. Meanwhile, audits of these geographically dispersed MNCs are challenging, and parts of these audits are sometimes outsourced to local foreign auditors who have no experience serving as opinion-issuing auditors for U.S. issuers (“inexperienced component auditors”). In this article, we examine whether involving such inexperienced component auditors is associated with MNCs’ internal information environment. We find that, relative to a matched control sample, Securities and Exchange Commission (SEC) issuers with the participation of inexperienced component auditors have higher internal information asymmetry and lower internal capital allocation efficiency. Our results suggest that audit quality matters not only to a firm’s external financial statement users but also to its internal users.

Keywords

component auditor internal information asymmetry internal capital allocation efficiency audit quality group audits

Introduction

With the global expansion of business operations, multinational corporations’ (MNCs) knowledge of their own operations is often dispersed across numerous divisions, which could lead to heightened internal information asymmetry (IIA) between parent and subsidiary entities (Chen et al., 2017). The quality of internal information flow from lower-level divisions and subsidiaries to top corporate managers ultimately affects the information sets top managers have for corporate-level decision-making. Considering a company’s external communication with the capital markets, misstatements detected and corrected in the audit process improve reporting quality and increase the information usefulness to external investors and stakeholders. In a similar vein, if an auditor detects misstatements at the subsidiary level, the correction improves information quality and could mitigate IIA, especially for geographically dispersed MNCs with large hierarchies. Using principal auditors’ characteristics (e.g., industry expertise and audit firm resources), prior studies have extensively examined how auditors with higher quality provide informational advantages to their clients (e.g., Bae et al., 2017). However, little is known about how subsidiary audit quality affects a company’s internal information environment, presumably due to the lack of data on subsidiary audit quality.

In this article, we use a unique hand-collected sample to identify subsidiary auditors participating in the group audits of conglomerates (i.e., audits in which multiple auditors participate). We collect these data from audit firm annual reports (Public Company Accounting Oversight Board [PCAOB] Form 2 filings), which identify audit firms that “did not issue audit reports on issuers, but played a substantial role in the preparation or furnishing of audit reports with respect to an issuer” (PCAOB, 2017). Thus, other than limited experiences as participating component auditors, these auditors have no experience auditing U.S. publicly traded companies as principal auditors (we label them as “inexperienced component auditors”).¹ Using the same hand-collected data, Dee et al. (2015) find that audit quality (both perceived and actual) is significantly lower for issuers that have been disclosed as having such inexperienced component auditors involved in their audits. Following Dee et al. (2015), we match each of our inexperienced component auditor observations with a control issuer using principal auditor and the proportion of foreign revenue as the matching criteria. Matching by principal auditor allows us to isolate the incremental effect of involving an inexperienced component auditor, whereas matching by percentage of foreign revenue allows us to identify comparable MNC controls. Using this matched sample, we examine whether involving an inexperienced component auditor, which presumably conducts lower-quality audits as suggested by Dee et al. (2015), affects MNCs’ internal information environment.

Measuring IIA is challenging, as it is not directly observable to outsiders. Chen et al. (2017) develop a novel measure to capture IIA within conglomerate firms using the difference in trading profits between top corporate managers and divisional managers on their own company’s stock, that is, divisional managers’ trading profits less top managers’ trading profits (DIFRET). Typically, top managers enjoy an information advantage over divisional managers because they can aggregate information from multiple business units and translate that into meaningful company-level trends and patterns. Indeed, Chen et al. (2017) document that values for DIFRET are generally negative (i.e., top managers earn higher returns than divisional managers). However, as Chen et al. (2017, p. 72) point out, “division-level information can be soft in nature and, therefore, difficult to transmit to headquarters. . . Incentives due to career concerns and internal competition for resources can also motivate divisional managers to distort or withhold information from top management.” If internal information flow from lower-level business units is poor, top managers’ information advantage diminishes and, as a result, the value of DIFRET increases, indicating a higher IIA between top corporate managers and divisional managers. Following Chen et al. (2017), we use DIFRET to measure an issuer’s IIA and expect IIA to be higher when part of the audit is outsourced to an inexperienced component auditor. Consistent with our expectation, we find that SEC issuers with inexperienced component auditors are associated with higher IIA.

Next, we examine internal capital allocation decisions made by a firm’s top managers. Prior research documents that investment efficiency is positively associated with financial reporting quality (Bae et al., 2017; Biddle et al., 2009; Biddle & Hilary, 2006; Cheng et al., 2013; McNichols & Stubben, 2008; Shroff, 2016). These studies, however, largely focus on the level of investment and usually define the deviation from the expected investment (i.e., over- and underinvestment) as inefficiency in investment. Our study is more in line with the internal capital market literature in finance and management and examines how top corporate managers allocate firm resources to projects and subsidiaries. Firm capital is finite, and top managers use information regarding projects’ potential profitability to make allocation decisions. The quality of these allocation decisions depends on the quality of the information used. Information transparency within the firm, therefore, is paramount (D’Mello et al., 2017). We posit that a higher-quality subsidiary audit improves the information quality at the subsidiary level, which in turn facilitates top managers’ internal capital allocation decisions. Thus, we expect that issuers with inexperienced component auditors are associated with less efficient internal capital allocations.

A major obstacle to examining internal capital allocation efficiency has been the lack of data, since internal decisions made by firms are usually proprietary. A notable exception is the capital allocation decisions in multisegment companies that are required to disclose their sales and investments at the segment level. Following Cho (2015), we measure internal capital allocation efficiency as the weighted average of signed, segment-level capital expenditure deviations. Consistent with our expectation, we find that internal capital allocations are less efficient for issuers that have inexperienced component auditors involved in their audits. Finally, in further analyses, we find that our measures of excess firm value and accounting performance are negatively associated with the use of an inexperienced component auditor and that our main results are robust to controlling for geographic distance between firm headquarters and subsidiary locations.

Much of our article studies the association between the involvement of inexperienced component auditors and the internal information environment within conglomerate firms. We find that IIA tends to be higher and the internal capital allocation tends to be less efficient for issuers that have inexperienced component auditors involved in their audits. However, because the decision to have a component auditor and the selection of an inexperienced component auditor are not exogenously determined, our research design does not test for the causal effects of randomly having an inexperienced component auditor, and thus the results should be interpreted with caution. The question of why and how an inexperienced auditor is chosen to participate in the group audit is intriguing; it would contribute significantly to the group audit literature and should be of interest to academics and regulators. While formal investigation of this question is beyond the scope of this study, we provide an in-depth discussion in the “Discussion and Future Research” section.

Our study makes two main contributions. First, prior research documents that high-quality external auditors improve the quality of information provided to a firm’s external financial statement users (Blay et al., 2011; Dhaliwal et al., 2011; Lennox & Pittman, 2011; Minnis, 2011). However, as summarized in Online Appendix Figure OA1, these studies focus on opinion-issuing auditors at the consolidated reporting level and the information provided to a firm’s external users. In contrast, we examine the role of component (as opposed to principal) auditors and explore whether audit quality is associated with internal (as opposed to external) information quality. Thus, our study extends and complements the audit research by documenting that audit quality is also associated with the quality of information available to top managers regarding their own firms’ internal operations.

Second, our study is related to a growing literature that ties auditing to corporate investment (e.g., Bae et al., 2017; Dhaliwal et al., 2016; Kausar et al., 2016). For example, Bae et al. (2017) find that auditor characteristics that proxy for an auditor’s knowledge and resources are positively associated with client investment efficiency. One key source of an auditor’s knowledge resources, as highlighted in Bae et al. (2017), is related to the industries in which the clients operate, that is, an auditor’s industry expertise. Their results suggest that auditors provide “informational advantages to their clients in a generalized investment setting” (Bae et al., 2017, p. 19). Dhaliwal et al. (2016), on the other hand, look at a specific setting of merger and acquisition (M&A) transactions. They find that when audit firms provide audit services to both the target and the acquirer prior to the acquisition, the presence of shared auditors is associated with better M&A transaction outcomes, suggesting that shared auditors facilitate the flow of information from the target to the acquirer. Our article differs in that prior studies focus on auditors’ ex ante possession of superior information (e.g., their knowledge about a client’s industry or target), whereas we are interested in whether the auditing procedures per se are associated with the quality of information supplied by the subsidiaries, although both could alter the information sets that top managers have for decision-making. In addition, Bae et al. (2017), as in many studies on investment efficiency, focus on the level of investment and examine over- and underinvestment issues (i.e., the size of the overall pie). Our study is one of the first to look at the role of auditing from the internal resource allocation perspective (i.e., the cutting of the pie).

Literature Review and Hypotheses Development

The Information Problems Within a Conglomerate

A conglomerate can be viewed in the principal-agent framework, with the parent entity representing the principal and the component entities (e.g., subsidiaries) representing agents (Shroff et al., 2014).² Agency problems within the firm may arise because the agent (e.g., the divisional or subsidiary managers) may not always act in the best interest of the principal (Jensen & Meckling, 1976). In a diversified firm, the parent entity allocates the firm’s resources to its subsidiaries, and then subsidiary managers invest those resources (Gertner et al., 1994). Top managers set and communicate the firm’s overall strategic objectives and are, therefore, well versed in the firm’s overall plans. However, divisional and subsidiary managers execute these plans and, unlike the parent entity’s top managers, have ready access to detailed operational information, such as local product demand and customer relationships (e.g., Bernardo et al., 2004; Chen et al., 2017). Division-level information like this is valuable to top management, and “the lack of free-flowing information from divisional managers to corporate headquarters constrains top management’s ability to accurately assess their firms’ performance, financial health and future prospects” (Chen et al., 2017, p. 72). Numerous factors, however, such as career concerns and internal competition for resources, may incentivize divisional managers to distort or withhold information from top management, resulting in IIA between a conglomerate’s top and divisional managers. When it comes to MNCs that have foreign operations, cross-border friction arising from “geographic dispersion, cultural and language differences, differing legal systems, etc.” may exacerbate the moral hazard and information friction (Shroff et al., 2014).

Prior research provides evidence related to the information problems that exist between top and divisional managers. In poor internal information environments, divisional managers tend to rent seek and use bargaining power to obtain greater capital budgeting allocations (Scharfstein & Stein, 2000). Information problems lead to the creation of managerial contracts, but these contracts sometimes result in corporate headquarters underinvesting in subsidiaries, especially when “the asymmetric information between headquarters and division managers is greater” (Bernardo et al., 2004). Furthermore, in high information asymmetry environments, CEOs resort to allocating capital based on social connections with divisional managers (Duchin & Sosyura, 2013). Conversely, firms with stronger internal information environments are better able to secure financial benefits, such as lower effective tax rates (Gallemore & Labro, 2015). Collectively these studies suggest that it is important for top managers to have high-quality financial information regarding subsidiary operations, especially for diversified firms.

Group Audits

Formally, a group audit is the audit of financial statements that include more than one component, where the term “component” is defined as an “entity or business activity” that is included in an issuer’s financial statements (American Institute of Certified Public Accountants (AICPA), 2016). Examples of components include subsidiaries, geographic and business segments, divisions, and investments (Westervelt, 2014). While an opinion-issuing auditor audits a conglomerate’s financial statement accounts at the consolidated level, a secondary component auditor often performs audit procedures over subsidiary financial statement accounts. For example, in 2014 Lovelock & Lewes, an audit firm located in India, audited a subsidiary of the company Cognizant Technology Solutions Corporation, while the audit firm PricewaterhouseCoopers LLP issued the audit opinion for the consolidated entity (Lovelock & Lewes, 2015).

In December 2015, the PCAOB issued Release No. 2015-008, which (beginning in 2017) requires disclosure “with respect to all other accounting firms that participated in the audit” (PCAOB, 2015).³ Before issuing the final rule, the PCAOB solicited comments. Many commenters opposed the requirement, arguing that such disclosures “were not useful information” or would confuse financial statement users about “the degree of responsibility for the audit assumed by the accounting firm signing the auditor’s report” (PCAOB, 2015). However, the PCAOB ultimately concluded that the disclosure of participating auditors was necessary given that companies have become increasingly global. In addition, PCAOB inspections had revealed that in some cases participating auditors complete “most of the audit work (or, in extreme cases, substantially all of the work)” (PCAOB, 2015). The board further noted “the quality of the audit is dependent, to some degree, on the competence and integrity of the participating accounting firms” (PCAOB, 2015).

In summary, there is information asymmetry between a conglomerate’s top and divisional managers, especially for geographically dispersed MNCs with large hierarchies. Top managers rely on accounting information to make operational decisions. Prior studies have documented evidence of IIA in conglomerates, which suggests that the quality of subsidiary operational information available to top managers is important. It is not yet known, however, if audit quality is associated with this asymmetry. Audits of these companies are referred to as group audits and often involve the participation of subsidiary auditors. Therefore, we argue a study of the effects of audit quality on a conglomerate’s internal information environment is best focused on subsidiary audit quality. We use the setting of group audits and measure audit quality by the experience level of a diversified firm’s subsidiary auditor.

Hypotheses Development

No study, to our knowledge, has examined how audit quality affects a firm’s internal information environment. The studies that are most directly comparable to our own are D’Mello et al. (2017) and Weber and Zheng (2017), who find that a firm’s internal control failures are negatively associated with internal capital allocation efficiency. In general, there is also evidence that audits in general of a firm’s consolidated financial statements improve a firm’s external information environment. For example, firms that voluntarily submit to an audit experience credit rating upgrades and lower costs of debt (Lennox & Pittman, 2011; Minnis, 2011). Evidence suggests that shareholders value the information provided by modified audit opinions (e.g., Blay et al., 2011). There is also evidence that when an auditor detects a material weakness in a firm’s internal control system, the firm’s credit spread increases (Dhaliwal et al., 2011). Taken together, this evidence suggests that a high-quality audit of a firm’s consolidated financial statements helps improve the quality of information provided to external financial statement users.

Similarly, there is evidence that high-quality, consolidated financial statement auditors improve a firm’s management information environment. For instance, Dhaliwal et al. (2016), in a sample of M&A transactions, find evidence consistent with the presence of a common auditor (i.e., an auditor shared by the acquirer and target firms) facilitating the flow of target company information to acquirer managers. Bae et al. (2017) find evidence consistent with industry expert auditors providing useful industry-level information to their clients. Together, these studies suggest that a high-quality consolidated financial statement auditor improves the quality of information available to top managers regarding the firm’s competitors, industry trends, and potential acquisition targets.

Finally, there is evidence in the group audit literature that the participation of component auditors, in the audit of a conglomerate’s subsidiary, affects overall audit quality and the quality of information provided to external financial statement users. For example, Dee et al. (2015) find that the participation of component auditors is associated with higher performance-adjusted, absolute discretionary accruals. Carson et al. (2016) find, for a sample of Australian multinational enterprises, that audit quality improved for group audit engagements following a strengthening of group audit auditing standards. Finally, Glover and Wood (2014) find that when a subsidiary’s financial statements are separately filed with the SEC and are included in another firm’s consolidated financial statements, subsidiary financial statement quality is higher.⁴ Together, these studies suggest that the quality of audit procedures performed over subsidiary financial statement accounts affects the quality of information provided to a conglomerate’s external financial statement users.

In a similar vein, we posit that subsidiary audit quality is associated with the quality of information available to top managers regarding the firm’s own operations. As described above, a component auditor assists a consolidated, opinion-issuing auditor in the audit of group financial statements. Some auditors serve exclusively as component auditors, while others sometimes serve (in separate audit engagements) as opinion-issuing auditors. We argue that the former type—a component auditor that participates in group audits but does not issue audit opinions for any U.S. publicly traded companies—is less experienced than the latter type. Ultimately, the opinion-issuing auditor is legally responsible for the audit of an issuer. Component auditors, on the other hand, have less responsibility; this smaller amount of legal responsibility has a negative effect on audit quality (Jia & Li, 2016). When an inexperienced component auditor participates in a group audit, the quality of the audit procedures performed over subsidiary financial statement accounts is likely to be lower, and subsidiary misstatements and internal control issues may go undetected. Following this line of reasoning, we predict that there will be a positive association between a firm’s IIA and the presence of an inexperienced component auditor.

Stated formally, our first hypothesis is as follows (alternative form):

Hypothesis 1: IIA is positively associated with the participation of inexperienced component auditors in a firm’s audit engagement.

Next, a key issue faced by the top managers of a multinational corporation is how best to allocate capital to subsidiaries. Prior research documents that investment efficiency is positively associated with financial reporting quality. For instance, Biddle and Hilary (2006) find that firms’ capital investments are less sensitive to cash flows when accounting quality is greater. Furthermore, financial reporting quality constrains (encourages) investment in cash-rich and unlevered (cash-poor and levered) firms (Biddle et al., 2009). Cheng et al. (2013) find that firms with internal control issues underinvest (overinvest) when they are financially constrained (unconstrained). In addition, managers make less efficient capital allocations following a restatement (McNichols & Stubben, 2008) and alter their investment decisions following changes in GAAP (Shroff, 2016). Bae et al. (2017) find that fixed asset investment efficiency is positively associated with auditor industry expertise and size. Bae et al., however, do not consider whether the participation of inexperienced subsidiary auditors is associated with internal capital allocation efficiency (i.e., a firm’s allocation of limited capital across subsidiaries). We conjecture that the participation of inexperienced component auditors will result in poor-quality subsidiary financial information. Consequently, we predict that there will be a negative association between a firm’s internal capital allocation efficiency and the presence of an inexperienced component auditor.

Stated formally, our second hypothesis is as follows (alternative form):

Hypothesis 2: Internal capital allocation efficiency is negatively associated with the participation of inexperienced component auditors in a firm’s audit engagement.

Research Design

Measure of IIA: DIFRET

The information asymmetry between top managers and divisional managers is typically hard to quantify and unobservable to researchers. Chen et al. (2017) develop a novel measure of IIA, which compares the trading profitability of top managers and divisional managers on their own company’s stock. Prior studies argue that the difference in the profitability of insider trades between two inside parties captures the difference in their private information sets (e.g., Ravina & Sapienza, 2010).⁵ The underlying assumption is that corporate insiders trade on their private information of the firm’s consolidated financial performance, and their trading profits, to some extent, reflect the quality of the information they possess. We focus on IIA between top executives and divisional managers. When IIA is low, high-quality information flows from divisional managers to the corporate headquarters. Once all the pieces of information get aggregated, top managers gain a superior knowledge about the firm’s consolidated financial performance. Indeed, the signed difference in the profitability of insider trades between divisional managers and top managers tends to be negative on average, suggesting that the superior knowledge allows top managers to trade more profitably (Chen et al., 2017). When IIA is high, divisional managers possess private information about their own business units, but the division-level information is withheld or distorted to some degree when it is transmitted to the headquarters due to various within-firm agency problems (Jensen & Meckling, 1976). While top managers still get information from all the divisions and can aggregate the information to assess their firm’s consolidated performance and future prospects, the building blocks (i.e., information from numerous divisions and business units) are of low quality. This could undermine top managers’ information advantage and reduce their trading profits, resulting in a smaller difference in the trading profitability between divisional and top managers (e.g., less negative). The lack of free-flowing high-quality information may even put top managers at an informational disadvantage relative to divisional managers, resulting in a positive signed difference in the trading profitability. Following this rationale, we expect that the degree of IIA between divisional managers and top managers will manifest in the difference in their trading profitability.

To construct DIFRET, we first obtain all the trades executed by corporate insiders during the current fiscal year from Thomson Reuters’ Insider Filing Data Feed. Then, we identify and exclude trades that are likely driven by liquidity needs and other routine reasons. Specifically, following Cohen et al. (2012), we examine insiders’ trading pattern and label the transactions as “routine” trades if an insider makes open-market trades in the same calendar month over a period of at least three consecutive years. These trades are unlikely to capture insiders’ private information and thus are excluded. In contrast, the remaining nonroutine (or “opportunistic”) trades likely reflect managers’ incentive to take advantage of their private information.

Following Chen et al. (2017), top managers include company executives with the following titles: chairman (Thomson Reuters role code = CB), vice chairman (VC), CEO, CFO, and Chief Operating Officer (CO), whereas divisional managers include Divisional Officers (OX), Officer of Subsidiary company (OS), and other non–top executives with role codes of AV, EVP, O, OP, OT, S, SVP, VP, GP, LP, M, MD, OE, TR, GM, C, and CP. We construct two variables, TOP_RET and DIV_RET. TOP_RET (DIV_RET), which captures the trading profits of top (divisional) managers for firm i in year t, is measured as the average cumulative abnormal return over the 6-month period after their opportunistic trades during the current fiscal year. We measure these returns in the same period during which an inexperienced component auditor participated in the audit of the issuer. DIFRET is measured as the difference between TOP_RET and DIV_RET (DIFRET = DIV_RET –TOP_RET). Larger values of DIFRET indicate higher IIA, in which case top managers’ information advantage over that of divisional managers is weaker.

Regression Model for the Analysis of IIA

We estimate the following model to test our Hypothesis 1:

\begin{matrix} DIFRE T_{it} = β_{0} + β_{1} \times InexpC A_{it} + β_{2} \times RO A_{it} + β_{3} \times MT B_{it} + β_{4} \times SIZ E_{it} \\ + β_{5} \times NUMSEGBU S_{it} + β_{6} \times NUMSEGGE O_{it} + β_{7} \times BIG 4 U S_{it} \\ + β_{8} \times BIG 4 FOREIG N_{it} + ε_{it} . \end{matrix}

(1)

The variable of interest is InexpCA, an indicator variable equal to one for issuers that have inexperienced component auditors as identified in the participating auditor’s Form 2, and zero otherwise. Consistent with Chen et al. (2017), the control variables include firm characteristics such as profitability (ROA), growth opportunities (MTB), the natural logarithm of firms’ market value of equity (SIZE), the number of issuer business segments (NUMSEGBUS), and the number of issuer geographic segments (NUMSEGGEO). In addition, we include two indicator variables to proxy for principal auditor quality (BIG4US and BIG4FOREIGN) following Dee et al. (2015). Under our first hypothesis, we expect the coefficient on InexpCA, β₁, to be positive.

Measure of Internal Capital Allocation Efficiency: ICAE

As was the case with IIA, it is difficult to observe or quantifiably measure internal capital market efficiency (ICAE). As Rajan et al. (2000, p. 35) point out, a major obstacle has been the lack of data because “data on internal decisions made by firms are generally proprietary.” A notable exception is the mandatory segment reporting, under which a diversified firm is required to disclose sales, profitability, and investments by major segments. Although segment-level data are not as fine as division- or subsidiary-level data, they have been widely used in the internal capital market literature.

To construct internal capital allocation efficiency (ICAE), we follow Cho (2015) and take the following three steps. First, a segment-level capital allocation CAPX deviation is calculated as the ratio of segment capital expenditures to firm capital expenditures minus the ratio of segment sales to firm sales. This approach considers the reinvestment of the segment’s own proceeds as passive and any capital allocation that is not proportionate to segment sales as active resource reallocation between segments. Second, a signed segment-level CAPX deviation is calculated as (+1) ×CAPX deviation if the segment q is greater than its sibling segments’ asset-weighted average q, and (–1) ×CAPX deviation if the segment q is not greater than its sibling segments’ asset-weighted average q. Thus, a signed segment-level CAPX deviation takes a more positive value if a segment with higher (lower) growth opportunities receives more (less) capital than what would be expected under passive capital allocation. Third, the firm-level measure of internal capital allocation efficiency, ICAE, is calculated by weight-averaging the signed CAPX deviation across all segments within a firm. ICAE with a more negative value indicates more value-decreasing transfers of capital from a segment with higher growth opportunities to a segment with lower opportunities.

Regression Model for the Analysis of Internal Capital Allocation Efficiency

We estimate the following model to test our Hypothesis 2:

\begin{matrix} ICA E_{it} = γ_{0} + γ_{1} \times InexpC A_{it} + γ_{2} \times SIZ E_{it} + γ_{3} \times MT B_{it} + γ_{4} \times LEVERAG E_{it} \\ + γ_{5} \times CAPE X_{it} + γ_{6} \times NonCAPE X_{it} + γ_{7} \times TANGIBILIT Y_{it} + γ_{8} \times CAS H_{it} \\ + γ_{9} \times DIVIDEN D_{it} + γ_{10} \times NUMSEGBU S_{it} + γ_{11} \times NUMSEGGE O_{it} \\ + γ_{12} \times DIVERSIT Y_{it} + γ_{13} \times BIG 4 U S_{it} + γ_{14} \times BIG 4 FOREIG N_{it} + ε_{it} . \end{matrix}

(2)

The variable of interest is InexpCA, an indicator variable equal to one for firms that are identified in PCAOB Form 2 filings, and zero for the matched issuers. Consistent with Cho (2015), the regression includes a group of control variables to account for the effects of firm characteristics such as growth opportunities (MTB), the natural logarithm of firms’ market value of equity (SIZE), financial constraints (LEVERAGE, CASH, and DIVIDEND), capital expenditure at the firm level (CAPEX and NonCAPEX), and asset mix (TANGIBILITY). In addition, the model includes segment characteristics such as the number of issuer business segments (NUMSEGBUS), the number of issuer geographic segments (NUMSEGGEO), and the segment industry diversity (DIVERSITY), and two proxies for principal auditor quality (BIG4US and BIG4FOREIGN). Under our second hypothesis, we expect the coefficient on InexpCA, γ₁, to be negative.

Sample and Results

Audit Firm Form 2 Filings

We hand-collect data on inexperienced component auditors from audit firm annual reports filed with the PCAOB (Form 2).⁶ As described in the “Literature Review and Hypotheses Development” section, we consider a component auditor to be inexperienced if it participates in the audit of but does not issue audit opinions for any publicly listed U.S. firms (i.e., the opinion-issuing auditor is legally responsible for the audit of an issuer; component auditors, on the other hand, have less responsibility; Jia & Li, 2016). We hand-coded component auditor information from Form 2 Items 1.1 (“Name of the Firm”), 1.2 (“Contact Information of the Firm”), and 4.2 (“Audit Reports with Respect to Which the Firm Played a Substantial Role during the Reporting Period”). We identified 1,350 inexperienced component auditor observations for years 2006 through 2016 and report the sample distribution by year and industry in Online Appendix Table OA1.

Sample for the IIA Analyses

Table 1, Panel A summarizes our sample selection process for our IIA analyses. After data requirements, we obtain 204 observations that have inexperienced component auditors involved in the auditing process. Next, we use a one-to-one matching process—our matching criteria included identifying firms with the same principal auditor and those with the closest percentage of foreign revenue as the test observation—to construct a control sample for which an inexperienced component auditor did not participate in the audit of an issuer (our matching procedure is consistent with Dee et al., 2015, p. 1948). This matching process results in 181 pairs of test and control firms, and it is meant to ensure that differences in audit quality, between a test observation and its matched control observation, are driven by component auditors and not consolidated, opinion-issuing auditors. Finally, Panel B of Table 1 provides frequency counts for our test sample. Of our 181 inexperienced component auditor observations, 121 (66.9%) specifically performed subsidiary audit procedures; the remaining 60 observations participated in a group audit but performed other functions. Finally, 72% (28%) of our firm-year observations had a Big 4 (non–Big 4) principal auditor.

Table 1.

Sample for Internal Information Asymmetry (IIA) Analyses.

Panel A: Sample Selection for IIA Analyses.
Issuers identified in PCAOB Form 2 filings of audit firms that “did not issue audit reports on issuers, but played a substantial role in the preparation or furnishings of audit reports with respect to an issuer” for years 2006–2016	1,350
Less: issuers that cannot be matched to Compustat	(296)
	1,054
Less: issuers with insufficient insider trading data to calculate IIA	(850)
	204
Less: issuers that fail to find a matching issuer (control sample)	(23)
Final sample for Internal Information Asymmetry (IIA) analyses	181

Panel B: Component Auditor’s Role in Issuer’s Audit for IIA Sample (N = 181).
	Principal auditor			Total
Component auditor’s role	U.S. Big 4(n = 127; 70%)	Foreign Big 4(n = 3; 2%)	Non–Big 4(n = 51; 28%)	Total
Audited issuer’s subsidiary	76	3	42	121
	59.8%	100.0%	82.4%	66.9%
Other	51	0	9	60
	40.2%	0.0%	17.6%	33.1%

Note. Panel A summarizes our sample selection process for our internal information asymmetry analyses. Panel B cross-tabulates the frequency of component auditor type with consolidated auditor type. “Audited issuer’s subsidiary” identifies component auditors who, in their PCAOB annual report (Form 2), state that they specifically audited an issuer’s subsidiary. “Other” identifies all remaining observations (e.g., served as a subcontractor assisting the principal auditor).

Results for the IIA Analyses

In Table 2, we begin our IIA analyses. Panel A provides t tests comparing our test group with our matched control group. We find evidence of greater information asymmetry (larger DIFRET) in the treatment group than in the control group. We also find our test sample has a larger number of geographic business segments than our control sample. Panel B provides descriptive statistics on the top 10 subsidiary countries appearing in our IIA sample. For example, 115 of the firms in our test sample (63.5%) and 116 of the firms in our control sample (64.1%) have subsidiaries operating in the United Kingdom. Overall, the descriptive evidence suggests we have a good balance in the distribution of countries across the test and control samples.

Table 2.

Descriptive Statistics for Internal Information Asymmetry (IIA) Analyses.

Panel A: IIA Descriptive Statistics.
	Form 2 sample(N = 181)		Matched sample(N = 181)		Diff. in means
Variable	M	Median	M	Median	Diff.	t-stat.
DIFRET	−0.009	0.003	−0.063	−0.016	0.055	1.63*
ROA	0.066	0.075	0.042	0.065	0.023	1.39
MTB	2.572	2.198	3.676	2.826	−1.104	−1.31
SIZE	7.641	7.404	7.333	7.145	0.308	1.50
NUMSEGBUS	2.061	1.000	1.928	1.000	0.133	0.72
NUMSEGGEO	5.066	4.000	4.116	4.000	0.950	4.08***

Panel B: The Top 10 Subsidiary Countries Appearing in the IIA Sample.
		Form 2 sample(N = 181)		Matched sample(N = 181)		Diff. in %
#	Country	n	%	n	%	Diff.	t-stat.
1	United Kingdom	115	63.5	116	64.1	−0.6	−0.11
2	Canada	108	59.7	112	61.9	−2.2	−0.43
3	Germany	113	62.4	100	55.2	7.2	1.39
4	China	94	51.9	106	58.6	−6.7	−1.27
5	Netherlands	95	52.5	93	51.4	1.1	0.21
6	Singapore	102	56.4	79	43.6	12.8	2.43**
7	France	85	47.0	95	52.5	−5.5	−1.05
8	Mexico	85	47.0	82	45.3	1.7	0.32
9	India	81	44.8	82	45.3	−0.5	−0.11
10	Japan	74	40.9	79	43.6	−2.7	−0.53

Note. This table presents descriptive statistics for internal information asymmetry (IIA) analyses. See Appendix for variable definitions. Note that in Panel A we exclude the variables BIG4US and BIG4FOREIGN because our matching procedure required that our test sample (“Form 2 Sample”) and our control sample (“Matched Sample”) have the same principal auditor; thus, there is no potential for difference between the test and control samples for these variables. In Panel B, we report descriptive statistics related to the top 10 subsidiary countries appearing in our IIA sample. ***, **, and * denote significance at the 0.01, 0.05, and 0.10 levels, respectively (two-tailed tests).

Table 3 presents the results of our multivariate tests of IIA. In column 1, we include the indicator variable InexpCA, which is coded 1 if an inexperienced auditor participated in the audit engagement of an issuer, coded 0 otherwise. In column 2, we separately examine the type of inexperienced component auditor. InexpCA_Subsidiary is coded 1 if the component auditor disclosed that it worked specifically on the audit of an issuer’s subsidiary, coded 0 otherwise. InexpCA_OtherRole is coded 1 if the component auditor disclosed that it performed some other, non-subsidiary-related, role in the audit of an issuer, coded 0 otherwise. Hypothesis 1 predicts a positive association between IIA and the participation of an inexperienced component auditor. Consistent with this prediction, InexpCA and InexpCA_Subsidiary are positive and significant in columns 1 and 2, respectively. In terms of economic significance, the 0.052 coefficient on InexpCA implies that having an inexperienced component auditor is associated with an increase in DIFRET by 5.2%. Recall that DIFRET is measured as divisional managers’ trading profits (DIV_RET) minus top managers’ trading profits (TOP_RET). As reported in Panel A of Table 2, the mean of DIFRET is −0.063 for firms without inexperienced component auditors, suggesting that top managers’ trading profitability is, on average, higher than that of divisional managers by 6.3%. Thus, a 5.2% increase in DIFRET suggests that having an inexperienced component auditor is associated with a higher IIA, a reduced informational advantage for top managers, and a 5.2% reduction in relative trading profitability for top managers. These results suggest that when an inexperienced auditor participates in a group audit—specifically, when an inexperienced auditor performs audit work for an issuer’s subsidiary—IIA is greater and the difference is both economically and statistically significant. Overall, the evidence presented in Tables 2 and 3 is consistent with Hypothesis 1.

Table 3.

Analyses of Internal Information Asymmetry (IIA).

		Dependent variable = DIFRET
Independent variable	Predicted sign	(1)	(2)
Constant		−0.065(−1.05)	−0.066(−1.05)
InexpCA	(+)	0.052**(1.72)
InexpCA_Subsidiary	(+)		0.054**(1.67)
InexpCA_OtherRole			0.049(1.29)
ROA		0.035(0.43)	0.034(0.41)
MTB		0.001(0.69)	0.001(0.66)
SIZE		−0.003(−0.25)	−0.003(−0.22)
NUMSEGBUS		0.004(0.37)	0.004(0.38)
NUMSEGGEO		0.003(0.41)	0.003(0.39)
BIG4US		−0.006(−0.14)	−0.006(−0.14)
BIG4FOREIGN		−0.016(−0.35)	−0.016(−0.35)
Observations		362	362
Adjusted R²		0.010	0.010

Note. This table presents results from estimating Model 1. See Appendix for variable definitions. Standard errors are heteroskedasticity robust.

***

, **, and * denote significance at the 0.01, 0.05, and 0.10 levels, respectively (one-tailed tests for signed predictions, two-tailed tests otherwise).

Sample for the Internal Capital Allocation Efficiency (ICAE) Analyses

We next move to our tests of internal capital allocation efficiency (ICAE). Table 4, Panel A summarizes our sample selection process for our ICAE analyses. As in our IIA analysis, we begin constructing our sample with the 1,350 inexperienced component auditor observations that we identified from audit firms’ Form 2 filings. We matched these data to Compustat, CRSP, and Audit Analytics to construct our dependent and control variables. After data requirements, we obtain 164 inexperienced component auditor observations. Next, similar to our IIA analyses, we identify a control sample by matching on the principal auditor and foreign revenue. The resulting final sample for ICAE analyses has 124 pairs of treated and control firms. Finally, Panel B of Table 4 provides frequency counts for our test sample. We note that a majority of the sample (77 observations; 62.1% of sample) specifically performed subsidiary audit procedures; the remaining observations participated in a group audit but performed other functions.

Table 4.

Sample for Internal Capital Allocation Efficiency (ICAE) Analyses.

Panel A: Sample Selection for ICAE Analyses.
Issuers identified in PCAOB Form 2 filings of audit firms that “did not issue audit reports on issuers, but played a substantial role in the preparation or furnishings of audit reports with respect to an issuer” for years 2006–2016	1,350
Less: issuers that cannot be matched to Compustat	(296)
	1,054
Less: issuers that have no segment data	(181)
	873
Less: issuers that are reported as single-segment firms	(472)
	401
Less: issuers with insufficient segment data to calculate ICAE	(237)
	164
Less: issuers that fail to find a matching issuer (control sample)	(40)
Final sample for Internal Capital Allocation Efficiency (ICAE) analyses	124

Panel B: Component Auditor’s Role in Issuer’s Audit for ICAE Sample (n = 124).
	Principal auditor			Total
Component auditor’s role	U.S. Big 4(n = 56; 45%)	Foreign Big 4(n = 27; 22%)	Non–Big 4(n = 41; 33%)	Total
Audited issuer’s subsidiary	43	6	28	77
	76.8%	22.2%	68.3%	62.1%
Other	13	21	13	47
	23.2%	77.8%	31.7%	37.9%

Note. Panel A summarizes our sample selection process for our internal capital allocation efficiency analyses. Panel B cross-tabulates the frequency of component auditor type with consolidated auditor type. “Audited issuer’s subsidiary” identifies component auditors who, in their PCAOB annual report (Form 2), state that they specifically audited an issuer’s subsidiary. “Other” identifies all remaining observations (e.g., served as a subcontractor assisting the principal auditor).

Results for the Internal Capital Allocation Efficiency (ICAE) Analyses

Table 5, Panel A compares our inexperienced component auditor sample (Form 2 sample) with our matched control sample. Consistent with Hypothesis 2, we find evidence that our treatment group has significantly lower internal capital allocation efficiency (ICAE) than the matched control group, suggesting that the participation of an inexperienced component auditor is negatively associated with internal capital allocation. Furthermore, we find that our test sample has less leverage (LEVERAGE), has higher levels of NonCAPEX, pays dividends less often (DIVIDEND), and has a greater number of geographic business segments (NUMSEGGEO) than our matched control sample. Panel B provides descriptive statistics on the top 10 subsidiary countries appearing in our sample. As with the IIA sample, the descriptive evidence suggests we have a good balance in the distribution of countries across the test and control samples.

Table 5.

Descriptive Statistics for Internal Capital Allocation Efficiency (ICAE) Analyses.

Panel A: ICAE Descriptive Statistics.
Variable	Form 2 sample(n = 124)		Matched sample(n = 124)		Diff. in means
	M	Median	M	Median	Diff.	t-stat.
ICAE	−0.056	−0.003	0.013	−0.010	−0.069	−1.56*
SIZE	6.670	6.956	6.427	6.651	0.242	0.77
MTB	1.988	1.492	1.997	1.533	−0.009	−0.03
LEVERAGE	0.505	0.532	0.571	0.548	−0.066	−2.32**
CAPEX	0.202	0.181	0.199	0.186	0.003	0.19
NonCAPEX	0.718	1.000	0.524	1.000	0.194	3.19***
TANGIBILITY	0.302	0.222	0.291	0.212	0.011	0.38
CASH	0.118	0.106	0.140	0.094	−0.022	−1.40
DIVIDEND	0.403	0.000	0.532	1.000	−0.129	−2.05**
NUMSEGBUS	3.766	3.000	3.742	4.000	0.024	0.12
NUMSEGGEO	5.250	4.500	4.040	3.000	1.210	3.20***
DIVERSITY	0.662	0.600	0.630	0.500	0.032	0.95

Panel B: The Top 10 Subsidiary Countries Appearing in the ICAE Sample.
		Form 2 sample(n = 124)		Matched sample(n = 124)		Diff. in %
#	Country	n	%	n	%	Diff.	t-stat.
1	United Kingdom	71	57.3	57	46.0	11.3	1.91*
2	Canada	67	54.0	59	47.6	6.4	1.01
3	China	60	48.4	57	46.0	2.4	0.38
4	Mexico	47	37.9	67	54.0	−16.1	−2.57**
5	France	53	42.7	40	32.3	10.4	1.71*
6	Brazil	46	37.1	45	36.3	0.8	0.13
7	Germany	48	38.7	43	34.7	4.0	0.66
8	Netherlands	43	34.7	48	38.7	−4.0	0.66
9	Australia	45	36.3	43	34.7	1.6	0.26
10	Singapore	43	34.7	45	36.3	−1.6	0.26

Note. This table presents descriptive statistics for internal capital allocation efficiency analyses. See Appendix for variable definitions. Note that in Panel A we exclude the variables BIG4US and BIG4FOREIGN because our matching procedure required that our test sample (“Form 2 Sample”) and our control sample (“Matched Sample”) have the same principal auditor; thus, there is no potential for difference between the test and control samples for these variables. In Panel B, we report descriptive statistics related to the top 10 subsidiary countries appearing in our ICAE sample. ***, **, and * denote significance at the 0.01, 0.05, and 0.10 levels, respectively (two-tailed tests).

Table 6 presents results from estimating our Model 2. As in our IIA analyses, we examine the effect of inexperienced component auditor (InexpCA) in column 1, and then separately examine the type of inexperienced component auditor (InexpCA_Subsidiary versus InexpCA_OtherRole) in column 2. Our Hypothesis 2 predicts a negative association between internal capital allocation efficiency and the participation of an inexperienced component auditor. Consistent with this prediction, we find that our dependent variable, ICAE, is negatively associated with InexpCA and InexpCA_Subsidiary in columns 1 and 2, respectively. We also find evidence that the presence of a high-quality principal auditor is positively associated with internal capital allocation efficiency (BIG4US and BIG4FOREIGN are both significantly positive in columns 1 and 2).

Table 6.

Analyses of Internal Capital Allocation Efficiency (ICAE).

		Dependent variable = ICAE
Independent variable	Predicted sign	(1)	(2)
Constant		−0.129(−0.75)	−0.087(−0.56)
InexpCA	(−)	−0.085*(−1.57)
InexpCA_Subsidiary	(−)		−0.124**(−1.83)
InexpCA_OtherRole			−0.018(−0.37)
SIZE		−0.049**(−2.21)	−0.053**(−2.29)
MTB		0.005(0.71)	0.006(0.80)
LEVERAGE		0.068(0.66)	0.055(0.54)
CAPEX		−0.134(−0.44)	−0.145(−0.47)
NonCAPEX		0.172*(1.92)	0.168*(1.92)
TANGIBILITY		0.310(1.41)	0.283(1.34)
CASH		−0.073(−0.35)	−0.100(−0.48)
DIVIDEND		0.117*(1.68)	0.122*(1.74)
NUMSEGBUS		0.007(0.82)	0.008(0.82)
NUMSEGGEO		0.006(1.01)	0.006(1.06)
DIVERSITY		0.058(0.60)	0.051(0.55)
BIG4US		0.177**(2.07)	0.190**(2.13)
BIG4FOREIGN		0.167*(1.94)	0.154*(1.84)
Observations		248	248
Adjusted R²		0.139	0.148

Note. This table presents results from estimating Model 2. See Appendix for variable definitions. Standard errors are heteroskedasticity robust.

***

, **, and * denote significance at the .01, .05, and .10 levels, respectively (one-tailed tests for signed predictions, two-tailed tests otherwise).

Additional Analyses

Diversification Discount and Accounting Performance

Researchers in finance and strategy have documented a variety of empirical evidence that diversified firms trade, on average, at lower stock values than comparable portfolios of single-segment firms (e.g., Berger & Ofek, 1995; Comment & Jarrell, 1995; Lang & Stulz, 1994; Servaes, 1996). The lower valuation is typically referred to as “diversification discount” in the literature, and many attribute the discount to inefficient allocation of resources across segments (Berger & Ofek, 1995; Billett & Mauer, 2003; Rajan et al., 2000). In the previous section, we find less efficient internal capital allocation for firms with inexperienced component auditors. For completeness, we next explore whether the adverse impact of inexperienced auditors on internal capital markets manifests in a lower firm value.

We use the excess value measure developed by Berger and Ofek (1995) to proxy for the valuation effect of diversification. Excess value (EXVAL) compares a firm’s actual value with its imputed value of segments estimated from industry-matched stand-alone firms.⁷ If the internal misallocation of resources to divisions with poor opportunities creates value losses, we expect a negative coefficient on InexpCA in the following regression:

\begin{matrix} EXVA L_{it} = α_{0} + α_{1} \times InexpC A_{it} + α_{2} \times SIZ E_{it} + α_{3} \times RO A_{it} + α_{4} \times CAPE X_{it} \\ + α_{5} \times NUMSEGBU S_{it} + α_{6} \times NUMSEGGE O_{it} + α_{7} \times DIVERSIT Y_{it} \\ + α_{8} \times BIG 4 U S_{it} + α_{9} \times BIG 4 FOREIG N_{it} + ε_{it} . \end{matrix}

(3)

Furthermore, to obtain more direct evidence, we also examine the ultimate driver of firm value, that is, the operating performance as captured by accounting numbers. Similar to excess value, a firm’s adjusted operating performance (ADJ_ROA) is constructed by comparing the firm’s actual ROA with the weighted average ROA of all the segments, where the segment-level ROA is based on the median single-segment firm in the same industry.⁸ If firms with inexperienced component auditors are more likely to engage in inefficient cross-subsidization of failing business segments, we predict a negative association between ADJ_ROA and InexpCA in the following regression:

\begin{matrix} ADJ_RO A_{it} = δ_{0} + δ_{1} \times InexpC A_{it} + δ_{2} \times SIZ E_{it} + δ_{3} \times CAPE X_{it} + δ_{4} \times NUMSEGBU S_{it} \\ + δ_{5} \times NUMSEGGE O_{it} + δ_{6} \times DIVERSIT Y_{it} + δ_{7} \times BIG 4 U S_{it} \\ + δ_{8} \times BIG 4 FOREIG N_{it} + ε_{it} . \end{matrix}

(4)

We use the same matching criterion as was used in our main analyses, identifying control firms with the same principal auditor and the closest percentage of foreign revenue. We require both the test and control firms to be multisegment firms, because we aim to investigate, once diversified, whether and how the excess firm value and relative operating performance vary for firms with versus without inexperienced component auditors.⁹

Table 7 presents the results of estimating Models 3 and 4. In column 1, we find a significantly negative coefficient on InexpCA (−0.132 with a t-statistic of −2.53), suggesting that the value loss associated with diversification is significantly larger for firms with inexperienced component auditors. The lost value of 13.2% appears economically significant and comparable in magnitude to the average diversification discount of 14.4% that Berger and Ofek (1995) report for all the multisegment firms. In column 3, we find a negative association between ADJ_ROA and InexpCA, suggesting that the operating performance as captured by accounting numbers is also lower for the test firms. Further analyses in columns (2) and (4) show that the lower firm value and operating performance are mainly driven by firms that engage inexperienced auditors to audit subsidiaries. Together these results provide additional support that firms involving inexperienced component auditors tend to have higher IIA and less efficient internal capital markets, which manifest in lower contemporaneous accounting performance as well as a lower expectation about future prospects.

Table 7.

Analyses of Diversification Discount and Accounting Performance.

Independent variable	Predicted sign	Dependent variable = EXVAL		Dependent variable = ADJ_ROA
		(1)	(2)	(3)	(4)
Constant		−1.070***(−7.56)	−1.067***(−7.54)	−0.082***(−6.45)	−0.082***(−6.37)
InexpCA	(−)	−0.132***(−2.53)		−0.012***(−2.47)
InexpCA_Subsidiary	(−)		−0.161***(−2.66)		−0.015***(−2.89)
InexpCA_OtherRole			−0.085(−1.22)		−0.006(−0.86)
SIZE		0.144***(8.14)	0.143***(8.02)	0.015***(11.31)	0.015***(11.01)
ROA		0.164(0.49)	0.161(0.49)
CAPEX		0.339**(2.51)	0.336**(2.42)	0.100***(7.21)	0.101***(7.10)
NUMSEGBUS		−0.020(−0.93)	−0.021(−0.95)	0.002(1.05)	0.002(1.11)
NUMSEGGEO		−0.019**(−2.45)	−0.018**(−2.36)	−0.001(−1.04)	−0.001(−1.04)
DIVERSITY		0.100(0.93)	0.115(1.05)	−0.027***(−2.89)	−0.026***(−2.72)
BIG4US		0.024(0.33)	0.028(0.37)	−0.014**(−1.98)	−0.013*(−1.96)
BIG4FOREIGN		0.203**(2.14)	0.190*(1.96)	−0.034***(−4.78)	−0.035***(−4.90)
Observations		456	456	654	654
Adjusted R²		0.284	0.284	0.279	0.279

Note. This table presents results from estimating Models 3 and 4. See Appendix for variable definitions. Standard errors are heteroskedasticity robust.

***

, **, and * denote statistical significance at the 0.01, 0.05, and 0.10 levels, respectively (one-tailed tests for signed predictions, two-tailed tests otherwise).

Geographic Distance

Although we match each of our inexperienced component auditor observations with a control firm on principal auditor and the percentage of foreign revenue, we recognize that our test and control groups may differ on the geographic distance between the headquarters and its subsidiaries, which could potentially affect a firm’s IIA.¹⁰ To address this concern, we develop a measure, GEODISTANCE, to capture the average geographic distance between a sample firm’s headquarters and its subsidiaries.¹¹ We then add this variable to our DIFRET and ICAE regression models as an additional control. We find the coefficient on GEODISTANCE to be statistically insignificant in both models (untabulated). Our main results are robust to controlling for geographic distance. These results indicate that our matching procedures effectively identify comparable control firms and that the geographic distance between the headquarters and its foreign operations does not appear to be a correlated omitted variable.

Discussion and Future Research

Our study is closely related to an emerging literature that examines the use and characteristics of component auditors in U.S. multinational audits. Burke et al. (2020) report that audit work performed by less competent component auditors (i.e., who lack requisite experience in the client’s industry) is associated with adverse audit outcomes, such as higher likelihood of misstatement and higher likelihood of nontimely reporting. Krishnan et al. (2020) find that cost of debt is greater when group audits involve low-quality component auditors (i.e., who face PCAOB disciplinary orders or have negative inspection findings). These findings are consistent with the earlier study by Dee et al. (2015), who find that involvement of inexperienced component auditors is associated with lower audit quality and negative market reactions. Results of these studies suggest, overall, that certain component auditor characteristics (e.g., lack of experience and competence) are associated with various adverse outcomes. These findings, together with what we document in our study, raise the question of why these inexperienced or less competent auditors are selected to participate in the group audits. This is an important question—It not only pertains to our study but also contributes significantly to the group audit literature and should be of interest to academics and regulators.

Related to this question, several recent studies examine the determinants of component auditor use. Those studies find that component auditor use is determined by client size, complexity, and the extent of foreign operations (e.g., Burke et al., 2020; Docimo et al., 2021). This finding is consistent with practitioner statements that the use of component auditors is often unavoidable for clients with complex multinational operations (e.g., Downey & Westermann, 2021). While these emerging studies provide valuable insights on whether to involve any component auditor, the question of which component auditor to use is still largely unanswered.

Given that this question is so far unexplored in the archival research, we interviewed practitioners and explored survey research to gain institutional background and provide some initial thoughts on this issue. For instance, who is involved in selecting the component auditors? What are their incentives? What factors might influence component auditor selection? While our discussions with practitioners suggest that component auditors are often selected and proposed by the principal auditor and approved by the client’s audit committee, in some cases the audit committee can elect to engage a specific component auditor (Downey & Westermann, 2021), and in other situations the management team of the foreign subsidiary may appoint an auditor for that subsidiary (Carson et al., 2016; Sunderland & Trompeter, 2017). Overall, it seems that the principal auditor, the client company’s U.S. parent and foreign subsidiary management, and the audit committee interact with each other, and they all play an important role in selecting the component auditors.

There are many factors that could drive component auditor selection, some of which may explain why inexperienced or less competent audit firms may be used as component auditors. For example, (a) auditors familiar with U.S. GAAP, GAAS, and relevant SEC requirements (i.e., more experienced auditors) may not be available in the foreign location where a component audit needs to be performed;¹² (b) it may be cost-prohibitive to engage a better component auditor (perhaps the experienced auditors prioritize their own clients for which they are the signing auditors and therefore have limited resources to complete the component audit work in a timely manner for the fee offered; thus, an inexperienced auditor may be more cost effective); and (c) mandatory audit firm rotation rules in certain jurisdictions (e.g., Europe and South Africa) may limit the selection of foreign component auditors.

Another important factor that influences component auditor selection is globally networked firms (GNFs), whereby member firms are connected via cooperative agreements and operate under one brand (e.g., PwC-US, PwC-Brazil, and PwC-Zimbabwe).¹³ Principal auditors may prefer their in-network affiliate as the component auditor due to existing cooperative agreements, prior working relationship, lower coordination costs, or easier fee negotiations between the principal and component auditors. In addition, as Downey and Bedard (2019, p. 143) note, principal auditors may be “unwilling to propose use of an out-of-network firm to the client’s board for fear of providing other GNFs the opportunity to ‘win’ the engagement.” Thus, it seems that various reasons (other than the component auditor’s competence) may come into play such that an inexperienced in-network affiliate might be selected to perform the component audit work even though non-GNF member firms with more experience are available in the location of the foreign subsidiary.

On the other hand, the client’s foreign subsidiaries may have other considerations. For example, local statutes in foreign jurisdictions often require a stand-alone audit of local business operations (e.g., a wholly owned subsidiary of the multinational company); as a result, subsidiaries that require local statutory audits might choose their auditor independently of the parent firm (Downey & Bedard, 2019; Downey & Westermann, 2021). In this case, subsidiaries may select audit firms based on existing local relationships or the audit firms’ local expertise, rather than the auditors’ knowledge of U.S. standards. The selected audit firm therefore may perform the statutory audit of the subsidiary (to satisfy the local statutes) as well as the component audit work requested by the principal auditors. In addition, as Downey and Westermann (2021, p. 1419) point out, it is possible that the component auditors may act to “serve their own interests, or the interests of local management [of the subsidiary].” Thus, it is important to take into consideration how foreign subsidiaries and their local statutory laws influence the component auditor selection.

Finally, regulatory oversight and the client’s audit committees may play an important role. Due to concerns about component auditor quality, the PCAOB has taken several initiatives to standardize the oversight and evaluation of component auditors, one of which is to mandate disclosure of component auditors on PCAOB Form AP for audit reports signed after June 30, 2017. Subsequently, the Center for Audit Quality (CAQ) provided further guidance—Form AP Auditor Reporting of Certain Participants: A Tool for Audit Committees—to assist audit committee members in better understanding, evaluating, and overseeing audit participants as component auditors (Center for Audit Quality [CAQ], 2017). Overall, these regulatory initiatives and additional guidance indicate regulators’ intent to strengthen audit committee oversight, although it is not clear to what extent audit committees understand the role of component auditors and pay close attention to its selection before and after these initiatives.

In summary, the selection of component auditor may be a function of local audit market conditions (e.g., cost and availability of more experienced auditors in the subsidiary’s location, mandatory audit firm rotation rules in foreign jurisdictions), principal auditor characteristics (e.g., the availability of a network auditor and the trade-off between network affiliates versus out-of-network firms), client preference, and audit committee oversight. It is critical to understand why an inexperienced or less competent auditor is chosen to perform component audit work, and we encourage future research in this area.

Conclusion

In conducting audits of large multinational groups, principal auditors often engage other auditors to conduct parts of audits in distant locations (Sunderland & Trompeter, 2017). Regulators have raised significant concerns about the quality of multinational group audits (International Auditing and Assurance Standards Board [IAASB], 2015; PCAOB, 2016). Consistent with those concerns, Dee et al. (2015) report that audit quality (both perceived and actual) is significantly lower for issuers that have been disclosed as having inexperienced component auditors involved in their audits. In this study, we investigate whether involving an inexperienced component auditor, which presumably conducts lower-quality audits as suggested by Dee et al. (2015), affects MNCs’ internal information environment and internal capital allocation efficiency.

Using the difference in trading profits between top corporate managers and divisional managers as a proxy for IIA, we find that an issuer’s IIA tends to be higher when part of the audit is outsourced to an inexperienced component auditor. Furthermore, we find that firms involving inexperienced component auditors tend to have less efficient internal capital markets, which manifest in lower contemporaneous accounting performance as well as a lower expectation about future prospects. Finally, when we separately examine different types of inexperienced component auditors, we find that our main results are largely driven by firms that engage inexperienced auditors to audit subsidiaries. Taken together, our results shed light on the role of component auditors and how engaging inexperienced component auditors affects MNCs’ internal information environment and operational decision-making.

Supplemental Material

sj-pdf-1-jaf-10.1177_0148558X221116537 – Supplemental material for Inexperienced Component Auditors and the Internal Information Asymmetry of Multinational Corporations

Supplemental material, sj-pdf-1-jaf-10.1177_0148558X221116537 for Inexperienced Component Auditors and the Internal Information Asymmetry of Multinational Corporations by Tom Adams and Ying Zhou in Journal of Accounting, Auditing & Finance

Footnotes

Appendix

Appendix.

Variable Definitions.

Variable	Definition
DIFRET	Internal information asymmetry, measured as the abnormal return of divisional managers minus the abnormal return of top managers for insiders’ opportunistic trades executed during the fiscal year t.
ICAE	Internal capital allocation efficiency, constructed following Cho (2015). First, a segment-level CAPX deviation is calculated as [the ratio of segment capital expenditures to firm capital expenditures – the ratio of segment sales to firm sales]. Second, a signed segment-level CAPX deviation is calculated as (+1) × CAPX deviation if the segment q is greater than its sibling segments’ asset-weighted average q, and (–1) × CAPX deviation if the segment q is not greater than its sibling segments’ asset-weighted average q. Thus, a signed segment-level CAPX deviation takes a more positive value if a segment with higher (lower) opportunities receives more (less) capital than what would be expected under passive capital allocation. Third, the firm-level measure of capital allocation efficiency is calculated by weight-averaging the signed CAPX deviation across all segments within a firm.
EXVAL	Excess value, measured as the natural log of the ratio of a firm’s actual value to its imputed value. Actual firm value equals the sum of the market value of equity and the book value of total debt. We calculate the imputed value of each segment by multiplying the segment’s sales (sales multiplier) by the median ratio of market value to assets for single-segment firms in the same industry. The industry for each segment is defined as the narrowest SIC grouping that has at least five single-segment firms with at least $20 million of sales and sufficient data to calculate the market-to-assets ratio. The imputed segment values are summed to estimate the implied firm value.
ADJ_ROA	Adjusted return on assets, measured as a firm’s actual return on assets (ROA) minus its imputed ROA. A firm’s ROA equals its earnings before interest and taxes (EBIT) divided by the total assets. The imputed ROA of each segment equals the level of ROA for the median single-segment firm in the same industry. The industry for each segment is defined as the narrowest SIC grouping that has at least five single-segment firms with at least $20 million of sales and sufficient data to calculate ROA. The imputed firm ROA equals the sales-weighted average ROA of all the segments.
InexpCA	An indicator variable equal to one for firms that are identified in the participating auditor’s Form 2, and zero otherwise.
InexpCA_ Subsidiary	An indicator variable equal to one if the substantial role played by the participating auditor is “Audited Issuer’s Subsidiary” (as indicated on Form 2), and zero otherwise.
InexpCA_ OtherRole	An indicator variable equal to one if the substantial role played by the participating auditor is not “Audited Issuer’s Subsidiary” (as indicated on Form 2), and zero otherwise.
ROA	Earnings before interest and taxes (EBIT) divided by the total assets.
MTB	The book value of equity divided by the market value of equity.
SIZE	The natural logarithm of the firm’s market value of equity.
LEVERAGE	The total liabilities divided by the total assets.
CAPEX	The capital expenditures divided by the net property, plant, and equipment (PP&E).
NonCAPEX	An indicator variable equal to one if a firm reports a positive amount of research and development (R&D) or intangibles, and zero otherwise.
TANGIBILITY	The net PP&E divided by the total assets.
CASH	The sum of cash and cash equivalents divided by the total assets.
DIVIDEND	An indicator variable equal to one if a firm reports a positive amount of dividends for common stocks, and zero otherwise.
NUMSEGBUS	The number of business segments.
NUMSEGGEO	The number of geographic segments.
DIVERSITY	Segment industry diversity, measured as the number of segments with unique two-digit SIC codes divided by the total number of segments.
BIG4US	An indicator variable equal to one if the principal auditor is from a U.S. office of a Big 4 firm, and zero otherwise.
BIG4FOREIGN	An indicator variable equal to one if the principal auditor is from a Big 4 foreign affiliate, and zero otherwise.

Acknowledgements

We thank Kannan Raghunandan (associate editor), Bharat Sarath (associate editor), Xiao-Jun Zhang (editor), two anonymous reviewers, Frank Murphy, Michael Willenborg, and workshop participants at the University of Connecticut, the 2018 Deloitte/University of Kansas Auditing Symposium, the 2018 Temple Accounting Conference, and the 2018 AAA Annual Meeting for helpful comments. We also thank the University of Connecticut and La Salle University for financial support.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Tom Adams

Ying Zhou

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Tom Adams, Ph.D., CPA, CGMA is an Assistant Professor of Accounting at La Salle University in Philadelphia, PA. His research interests are in the areas of auditing, corporate governance, shareholder activism, and mergers and acquisitions (M&As).

Ying Zhou is an Assistant Professor of Accounting at the University of Connecticut. Her research interests include financial reporting, insider trading, and the role of gatekeepers in capital markets. She received her Ph.D. from the University of Florida.

References

AICPA. (2016). AU-C section 600: Special considerations—Audits of group financial statements (including the work of component auditors). http://www.aicpa.org/Research/Standards/AuditAttest/DownloadableDocuments/AU-C-00600.pdf

Bae

G. S.

Dhaliwal

D. S.

Lamoreaux

P. T.

(2017). Auditors and client investment efficiency. The Accounting Review, 92(2), 19–40.

Bens

Monahan

(2004). Disclosure quality and the excess value of diversification. Journal of Accounting Research, 42(4), 691–730.

Berger

Ofek

(1995). Diversification’s effect on firm value. Journal of Financial Economics, 37(1), 39–65.

Bernardo

A. E.

Cai

Luo

(2004). Capital budgeting in multidivision firms: Information, agency, and incentives. The Review of Financial Studies, 17(3), 739–767.

Biddle

Hilary

(2006). Accounting quality and firm-level capital investment. The Accounting Review, 81(5), 963–982.

Biddle

Hilary

Verdi

R. S.

(2009). How does financial reporting quality relate to investment efficiency? Journal of Accounting and Economics, 48(2–3), 112–131.

Billett

Mauer

(2003). Cross-subsidies, external financing constraints, and the contribution of the internal capital market to firm value. The Review of Financial Studies, 16(4), 1167–1201.

Blay

A. D.

Geiger

M. A.

North

D. S.

(2011). The auditor’s going-concern opinion as a communication of risk. Auditing: A Journal of Practice & Theory, 30(2), 77–102.

10.

Burke

Hoitash

(2020). The use and characteristics of foreign component auditors in U.S. multinational audits: Insights from Form AP disclosures. Contemporary Accounting Research, 37(4), 2398–2437.

11.

Carson

Simnett

Vanstraelen

Trompeter

(2016). Assessing initiatives to improve the quality of group audits involving other auditors [Working paper]. University of New South Wales Australia, Maastricht University, University of Central Florida.

12.

Center for Audit Quality. (2017). Form AP auditor reporting of certain participants: A tool for audit committees. https://www.thecaq.org

13.

CEPII. (2020). GeoDist. http://www.cepii.fr/cepii/en/bdd_modele/presentation.asp?id=6

14.

Chen

Martin

Roychowdhury

Wang

Billet

M. T.

(2017). Clarity begins at home: Internal information asymmetry and external communication quality. The Accounting Review, 93(1), 71–101.

15.

Cheng

Dhaliwal

D. S.

Zhang

(2013). Does investment efficiency improve after the disclosure of material weakness in internal control over financial reporting? Journal of Accounting and Economics, 56(1), 1–18.

16.

Cho

Y. J.

(2015). Segment disclosure transparency and internal capital market efficiency: Evidence from SFAS No. 131. Journal of Accounting Research, 53(4), 669–723.

17.

Cohen

Malloy

Pomorski

(2012). Decoding inside information. Journal of Finance, 67(3), 1009–1043.

18.

Comment

Jarrell

(1995). Corporate focus and stock returns. Journal of Financial Economics, 37, 67–87.

19.

Dee

C. C.

Lulseged

Zhang

(2015). Who did the audit? Audit quality and disclosures of other audit participants in PCAOB filings. The Accounting Review, 90(5), 1939–1967.

20.

Dhaliwal

D. S.

Hogan

Trezevant

Wilkins

(2011). Internal control disclosures, monitoring, and the cost of debt. The Accounting Review, 86(4), 1131–1156.

21.

Dhaliwal

D. S.

Lamoreaux

P. T.

Litov

Neyland

(2016). Shared auditors in mergers and acquisitions. Journal of Accounting and Economics, 61(1), 49–76.

22.

D’Mello

Gao

Jia

(2017). Internal control and internal capital allocation: Evidence from internal capital markets and multi-segment firms. Review of Accounting Studies, 22(1), 251–287.

23.

Docimo

Gunn

Michas

(2021). Do foreign component auditors harm financial reporting quality? A subsidiary-level analysis of foreign component auditor use. Contemporary Accounting Research, 38(4), 3113–3145.

24.

Downey

Bedard

(2019). Coordination and communication challenges in global group audits. Auditing: A Journal of Practice & Theory, 38(1), 123–147.

25.

Downey

Westermann

(2021). Challenging global group audits: The perspective of US group audit leads. Contemporary Accounting Research, 38(2), 1395–1433.

26.

Duchin

Sosyura

(2013). Divisional managers and internal capital markets. The Journal of Finance, 68(2), 387–429.

27.

Dyreng

(2020). Data and code: Exhibit 21 data. https://sites.google.com/site/scottdyreng/Home/data-and-code

28.

Ege

Kim

Wang

(2020). Do global audit firm networks apply consistent audit methodologies across jurisdictions? Evidence from financial reporting comparability. The Accounting Review, 95(6), 151–179.

29.

Gallemore

Labro

(2015). The importance of the internal information environment for tax avoidance. Journal of Accounting and Economics, 60(1), 149–167.

30.

Gertner

R. H.

Scharfstein

D. S.

Stein

J. C.

(1994). Internal versus external capital markets. The Quarterly Journal of Economics, 109(4), 1211–1230.

31.

Glover

S. M.

Wood

D. A.

(2014). The effects of group audit oversight on subsidiary entity audits and reporting [Working paper]. Brigham Young University.

32.

International Auditing and Assurance Standards Board. (2015). Staff audit practice alert: Responsibilities of the engagement partner in circumstances when the engagement partner is not located where the majority of the audit work is performed. http://archive.is/JTROB

33.

Jensen

M. C.

Meckling

W. H.

(1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360.

34.

Jia

(2016). An analysis of US accounting firms’ cross-country audit quality of China concepts stocks. China Journal of Accounting Studies, 4(2), 183–204.

35.

Kausar

Shroff

White

(2016). Real effects of the audit choice. Journal of Accounting and Economics, 62(1), 157–181.

36.

Krishnan

Mao

Zhang

(2020). Component auditors and lenders’ perception of audit quality [Working paper]. Bentley University, University of Texas at San Antonio, University of Colorado Denver.

37.

Lang

Stulz

(1994). Tobin’s Q, corporate diversification, and firm performance. Journal of Political Economy, 102(6), 1248–1280.

38.

Lennox

C. S.

Pittman

J. A.

(2011). Voluntary audits versus mandatory audits. The Accounting Review, 86(5), 1655–1678.

39.

Lovelock & Lewes. (2015). PCAOB form 2: Annual report form. https://rasr.pcaobus.org/Forms/FormSummary.aspx?ID=3D477985E7D3876A5618858F3DAD8C17

40.

Lundholm

R. J.

Rogo

Zhang

J. L.

(2014). Restoring the Tower of Babel: How foreign firms communicate with U.S. investors. The Accounting Review, 89(4), 1453–1485.

41.

Mayer

Zignago

(2011). Notes on CEPII’s distance measures: The GeoDist database [Working paper]. The French Centre for Research and Studies on the World Economy.

42.

McNichols

M. F.

Stubben

S. R.

(2008). Does earnings management affect firms’ investment decisions? The Accounting Review, 83(6), 1571–1603.

43.

Minnis

(2011). The value of financial statement verification in debt financing: Evidence from private U.S. firms. Journal of Accounting Research, 49(2), 457–506.

44.

Public Company Accounting Oversight Board. (2015). Release no. 2015-008: Improving the transparency of audits. https://pcaobus.org/Rulemaking/Docket029/Release-2015-008.pdf

45.

Public Company Accounting Oversight Board. (2016). Proposed amendments relating to the supervision of audits involving other auditors. https://pcaobus.org/News/Releases/Pages/PCAOB-proposal-other-auditors-4-12-16.aspx

46.

Public Company Accounting Oversight Board. (2017). Registration, annual and special reporting. https://rasr.pcaobus.org/Search/Search.aspx

47.

Rajan

Servaes

Zingales

(2000). The cost of diversity: The diversification discount and inefficient investment. Journal of Finance, 55(1), 35–80.

48.

Ravina

Sapienza

(2010). What do independent directors know? Evidence from their trading. Review of Financial Studies, 23(3), 962–1003.

49.

Scharfstein

D. S.

Stein

J. C.

(2000). The dark side of internal capital markets: Divisional rent-seeking and inefficient investment. The Journal of Finance, 55(6), 2537–2564.

50.

Servaes

(1996). The value of diversification during the conglomerate merger wave. Journal of Finance, 51(4), 1201–1225.

51.

Shroff

(2016). Corporate investment and changes in GAAP. Review of Accounting Studies, 22(1), 1–63.

52.

Shroff

Verdi

(2014). Information environment and the investment decisions of multinational corporations. The Accounting Review, 89(2), 759–790.

53.

Sunderland

Trompeter

G. M.

(2017). Multinational group audits: Problems faced in practice and opportunities for research. Auditing: A Journal of Practice & Theory, 36(3), 159–183.

54.

Weber

Zheng

(2017). Internal information quality and capital allocation within conglomerates [Working paper]. University of Connecticut.

55.

Westervelt

M. A.

(2014, February 18). The scoop on group audits: You may have them, even though you think you don’t. Journal of Accountancy. https://www.journalofaccountancy.com/news/2014/feb/20149529.html

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.23 MB