Corporate Reputation and Internal Control Quality: Evidence from Fortune 1000 Companies

Abstract

This paper examines the association between company reputation and internal control quality. The prior literature suggests that reputation concerns reduce the range of risky choices by management. Building on this idea, we propose that reputation concerns drive high-reputation firms to uphold strong internal control quality, leading to lower internal control risk as reflected by fewer material weaknesses in their internal controls. By analyzing Fortune 1000 companies, our study finds that high-reputation companies are motivated to safeguard their reputation, driven by their need to signal strong performance and by the monitoring pressure from high-quality auditors. As a result, these high-reputation companies are less likely to have internal control material weaknesses, reflecting lower internal control risk and higher internal control quality. Our study enhances the understanding of the role company reputation plays in corporate behavior and decision-making processes.

1. Introduction

Prior studies have shown that company reputation, conceptually referred as reputation concerns that affect players’ actions (Camerer & Weigelt, 1988), plays a significant role in various corporate and financial contexts. Specifically, researchers have demonstrated that a good company reputation contributes to firm value (Anderson & Smith, 2006; Filbeck & Preece, 2003), improves earnings quality (Cao et al., 2012), lowers financing costs (Diamond, 1991; Siegel, 2005), and enhances the chance to hire industry specialist auditors (Huang & Kang, 2018). Despite the growing body of research on company reputation, to our knowledge, no prior studies have directly examined the association between company reputation and internal control risk. Thus, we fill this theoretical gap by studying how a company’s reputation consideration affects its internal control risk, and therefore internal control quality.

Internal control quality serves as an important setting for examining the interaction between reputation consideration and internal control risk. Studies show that internal control quality is significantly associated with client firm characteristics (Doyle et al., 2007) and corporate governance mechanisms such as the qualification of CFOs and CFO turnover (Li et al., 2010). Additionally, internal control quality is an important input for the decision-making process by stakeholders. Studies show that client firms with private lenders, stronger boards, audit committees composed of more members possessing accounting expertise, highly educated internal audit functions, fewer auditor changes, and higher auditor independence tend to exhibit lower internal control risk and higher internal control quality (Hasan, 2021; Hoitash et al., 2009; Lin et al., 2011; Zhang et al., 2007).

Poor internal control quality, captured by more internal control material weaknesses, can increase the perceived risk of a company (Beneish et al., 2008; Lobo et al., 2020). With better internal control quality, high-reputation companies can lower perceived risks and reduce costs of capital (Diamond, 1991; Siegel, 2005). Therefore, high-reputation firms are motivated to ensure strong internal controls to minimize risk and safeguard their reputation. However, this prediction is not without tension. It is also possible that high-reputation firms have lower internal control quality as such firms can become complacent when they assume a good reputation can protect them from potential negative outcomes associated with lower internal control quality. Thus, it is unclear whether and how the corporate reputation affects a company’s internal control quality.

To examine the association between company reputation and internal control risk, and therefore internal control quality, we use reputation scores from Fortune’s America’s Most Admired Companies (MA) list as a measure of company reputation. The MAs are selected from the Fortune 1000 list (the 1000 largest U.S. companies ranked by revenue) and ranked based on survey results from executives, directors, and analysts across nine dimensions of reputation, including innovation, people management, use of corporate assets, social responsibility, quality of management, financial soundness, long-term investment value, quality of products/services, and global competitiveness. This list is widely recognized as a reliable measure of company reputation in academic research (Cao et al., 2015, p. 48). Following the prior literature, we measure internal control quality using material weaknesses in internal control reported by auditors under SOX 404 (b). We define an indicator, Internal Control Weakness (ICW), for client year observations in which material weaknesses are reported under SOX 404 (b). The higher a firm’s internal control quality, the less likely it is that material weaknesses are reported by auditors in a given year.

Using data from Fortune 1000 companies from 2014 through 2022, we find that higher-reputation companies are motivated to protect their reputation, driven both by their need to signal strong performances and by the monitoring pressure exerted by high-quality auditors. Consistent with the audit quality literature, we define high-quality auditors as those possessing greater expertise and stronger incentives to provide rigorous monitoring (DeAngelo, 1981; Francis, 2011). Consequently, these high-reputation companies are less likely to have internal control material weaknesses, reflecting higher internal control quality.

Our study makes several contributions to the existing literature. First, our study contributes to the accounting literature by investigating the association between company reputation and internal control risk. The prior literature has demonstrated that company reputation impacts firm value, earnings quality, financial reporting quality, financing cost, and auditors’ choices (Anderson & Smith, 2006; Cao et al., 2012; Diamond, 1991; Filbeck & Preece, 2003; Huang & Kang, 2018; Siegel, 2005). Additionally, a substantial body of research has examined how firm characteristics and corporate governance affect internal control quality (Doyle et al., 2007; Hoitash et al., 2009; Li et al., 2010; Zhang et al., 2007). However, the relationship between company reputation and internal control quality has not been previously addressed. Our paper provides direct evidence that company reputation significantly influences internal control quality.

Second, our study enhances the understanding of company reputation in auditing settings. Prior studies fail to examine the unique consequence of company reputation on client firms’ internal control quality, which was not clear ex ante. Our analyses indicate that higher-reputation companies are motivated to protect their reputation, due to their need to signal good performances and the monitoring pressure from high-quality auditors. As a result, these high-reputation companies are less likely to have internal control material weaknesses and thus have higher internal control quality.

The remainder of the paper is organized as follows. Section 2 provides the related literature and hypothesis development. Section 3 describes our research design. Section 4 presents the results and additional analysis. The Section 5 summarizes the conclusions.

2. Related Literature and Hypothesis Development

2.1. Determinants of Internal Control Quality

Following Public Company Accounting Oversight Board (PCAOB) Auditing Standard No. 5, internal control quality refers to the extent to which a company’s internal control over financial reporting is designed and operating effectively to prevent or detect material misstatements in a timely manner. A large body of research has examined the corporate-level factors that influence internal control quality. Doyle et al. (2007) examine the determinants of internal control deficiencies and find that material weaknesses in internal control are significantly associated with firm size, age, growth, financial position, and business complexity. Focusing on the disclosure of SOX Section 404 internal control weakness, Li et al. (2010) provide evidence that less qualified CFOs and higher CFO turnover are more likely linked to adverse SOX 404 opinions. Similarly, Masli et al. (2010) document that the implementation of internal control monitoring technology is associated with a lower likelihood of material weaknesses.

In addition, the prior literature also provides evidence on how corporate governance characteristics influence companies’ internal control quality. Hoitash et al. (2009) show that companies with stronger boards and audit committees that have more members with accounting and supervisory expertise tend to have a lower likelihood of disclosing a Section 404 material weakness. Similarly, Johnstone et al. (2011) find positive associations between internal control material weakness and the turnover of audit committees, board of directors, and top management.

Besides board characteristics, prior studies also examine the effect on internal control weakness by other internal and external monitoring mechanisms. For example, Lin et al. (2011) examine the association between internal audit and material weaknesses under SOX 404 and find that material weakness disclosures are negatively associated with the education level of the company’s internal audit function and the extent to which the internal audit function incorporates quality assurance techniques into fieldwork. Zhang et al. (2007) focus on the external auditor characteristics and present evidence that the disclosure of material weakness is positively associated with auditor changes and auditor independence. More recent research also highlights the moderating role of corporate governance. For example, Senan (2024) shows that corporate governance strengthens the relationship between internal audit quality and financial reporting quality, emphasizing the role of governance mechanisms in supporting reliable internal control and reporting systems.

In addition, recent studies continue to document the broader impact of internal control systems on firms’ operational and sustainability outcomes. Tao et al. (2023) find that strong internal control systems significantly improve firms’ environmental performance, while Liu et al. (2024) show that internal control quality facilitates sustainable practices and enhances overall firm performance. Taken together, the findings highlight that strong internal control structures play a crucial role in effective risk management and credible financial reporting.

Despite the vast amount of evidence documenting how firm characteristics and corporate governance influence internal control quality, there has been little evidence on the association between company reputation and internal control quality.

2.2. Company Reputation and Internal Control Quality

Following the prior literature, we define company reputation as “observers’ collective judgments of a corporation based on assessments of the financial, social, and environmental impacts attributed to the corporation over time” (Barnett et al., 2006, p. 34). As suggested by Barnett et al. (2006), reputation concerns motivate companies to take actions that provide long-term benefits rather than focusing on actions that favor their short-term interests. Building on prior accounting and economics research, we refer to high-reputation companies as firms with substantial reputational capital derived from stakeholders’ evaluations of their financial, social, and environmental performance (Barnett et al., 2006).

Prior studies have investigated the association between company reputation and corporate behavior. For example, Cao et al. (2012) find that higher-reputation companies report higher quality earnings using a large sample of Fortune 1000 companies. Similarly, Cao et al. (2015) document that higher-reputation companies enjoy a lower cost of equity financing. The reputation effect has also been shown to influence companies’ external auditor choices and audit fees (Huang & Kang, 2018, 2022).

More recent work further highlights the importance of reputation in corporate reporting, sustainability, and performance. Kodirjonova and Kim (2023) show that firms with stronger reputations exhibit a stronger positive relationship between corporate social responsibility disclosure (CSRD) and financial performance, emphasizing the governance and monitoring benefits embedded in reputation. Fu et al. (2024) also find that corporate social responsibility (CSR) reputation enhances firms’ operational efficiency during green and low-carbon transitions. Together, these studies provide additional evidence that company reputation motivates stronger accountability, transparency, and long-term value creation.

As suggested by prior research, high-reputation companies emphasize accountability, credibility, and trustworthiness, and tend to have greater incentives to protect their reputations. Amirkhani et al. (2024) argue that reputation concerns reduce the range of risky choices that managers consider and find evidence that high-reputation companies employ more conservative accounting practices due to the incentive to maintain strong reputation. Therefore, we expect that reputation concerns may motivate high-reputation firms to maintain high internal control quality, resulting in lower internal control risk and fewer internal control material weaknesses.

In this study, we investigate the association between company reputation and SOX Section 404 (b) material weaknesses, which are identified by auditors. Following the prior literature, we focus on 404 (b) weaknesses instead of 404 (a) weaknesses (internal control weaknesses identified by management) or SOX 302 weaknesses (disclosure control weaknesses) because 404 (b) disclosures are more accurate (T. Chen et al., 2024; Donelson et al., 2017). Our hypothesis is presented as follows:

H1. 

There is a negative association between company reputation and internal control material weakness.

3. Research Design

3.1. Measurement of Company Reputation

Following prior research (Amirkhani et al., 2024; Cao et al., 2012, 2015; Huang & Kang, 2018, 2022), we capture the company reputation of U.S. firms using Fortune’s America’s Most Admired Companies list from recent years. The Most Admired (MA) companies have been selected from the Fortune 1000 companies (the 1000 largest U.S. companies ranked by revenues) each year and are ranked based on a poll of executives, directors, and analysts on nine dimensions related to company reputation, including innovation, people management, use of corporate assets, social responsibility, quality of management, financial soundness, long-term investment value, quality of products/services, and global competitiveness. The list is considered one of the most widely used measures of company reputation in academic research (Cao et al., 2015, p. 48).

Each year, there are approximately 300 Fortune 1000 companies selected to be on the MA list. Although their reputation scores are published on Fortune’s website, only the scores in the most recent year are shown for companies on MA lists.1 Due to the limited variation in observed MA scores within firms as the data is collected, we capture the company reputation using an indicator variable (MA) constructed based on whether the company is selected to be on the MA list in a given year. Such measurement is also used in Cao et al. (2015). We also capture the company reputation with an alternative measure as a robustness check in our alternative analysis.

3.2. Measurement of Internal Control Quality

Section 404 (b) of the Sarbanes–Oxley Act (SOX 404 (b) hereafter), which has been effective for accelerated filers since 2004, mandates that auditors should report the effectiveness of a company’s internal controls. Compared with material weaknesses under SOX 404 (a) and SOX 302, material weaknesses disclosed under SOX 404 (b) are more accurate (T. Chen et al., 2024; Donelson et al., 2017). Following prior research (Y. Chen et al., 2017; T. Chen et al., 2024; DeFond & Lennox, 2017; Donelson et al., 2017; Liang et al., 2024), we capture the internal control quality of client firms using the internal control material weaknesses reported by auditors under SOX 404 (b) and define an indicator (ICW) for client year observations with material weaknesses reported under SOX 404 (b). The higher the internal control quality a client firm has, the less likely that internal control material weaknesses are reported by auditors in a given client year. In the alternative analysis, we replace the indicator variable with a count variable, which captures the total number of SOX 404 (b) material weaknesses, as a robustness check.

3.3. Model Specification

Following prior research (T. Chen et al., 2024; DeFond & Lennox, 2017; Donelson et al., 2017; Rice & Weber, 2012), we test our H1 on the association between company reputation and internal control material weakness using Equation (1).2

$$\begin{array}{cc}\hfill IC{W}_{i,t}={\beta }_0& + {\beta }_{1 }{MA}_{\mathrm{i},\mathrm{t}}+{\beta }_2 {Size}_{\mathrm{i},\mathrm{t}}+{\beta }_{3 }{Age}_{\mathrm{i},\mathrm{t}}+{\beta }_4 {Agg\text{\_}Loss}_{\mathrm{i},\mathrm{t}}+{\beta }_5 {Extreme\text{\_}Growth}_{\mathrm{i},\mathrm{t}}\hfill \\ & +{\beta }_6 {ROA}_{\mathrm{i},\mathrm{t}}+{\beta }_7 {D\text{\_}XFIN}_{\mathrm{i},\mathrm{t}}+{\beta }_{8 }{Agg\text{\_}M\&A}_{\mathrm{i},\mathrm{t}}+{\beta }_9{ Foreign\text{\_}Trans}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{10} {Segments}_{\mathrm{i},\mathrm{t}}+{\beta }_{11} {Big4}_{\mathrm{i},\mathrm{t}}+{\beta }_{12} {Restatement}_{\mathrm{i},\mathrm{t}}+{\beta }_{13} {Auditor\text{\_}Change}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{14} {Audit\text{\_}Fee}_{\mathrm{i},\mathrm{t}}+Year Fixed Effects+Industry Fixed Effects+u\hfill \end{array}$$

(1)

where i denotes client firm i and t denotes year t. We estimate Equation (1) using a Probit model. The dependent variable ICW is an indicator variable that is equal to one if the auditor identifies a SOX 404 (b) internal control material weakness in the year t for client i, and zero otherwise. The key variable of interest MA is an indicator that equals one if the firm is on the MA list in year t, and zero otherwise. We include several control variables based on prior research (T. Chen et al., 2024; DeFond & Lennox, 2017; Donelson et al., 2017; Doyle et al., 2007; Guo et al., 2016; Liang et al., 2024; Rice & Weber, 2012), including firm size (Size), firm age (Age), loss firm indicator (Agg_Loss), growth indicator (Extreme_Growth), return on assets (ROA), newly issued debt and equity (D_XFIN), M&A indicator (Agg_M&A), foreign transaction indicator (Foreign_Trans), number of segments (Segments), Big 4 auditor indicator (Big4), restatement indicator (Restatement), auditor change indicator (Auditor_Change), and audit fees (Audit_Fee). The definitions of all variables in Equation (1) are summarized in

Table 1. Variable definitions.

Variable		Definition
ICW	=	Internal control weakness indicator that equals one if the auditor identifies a SOX 404 (b) internal control weakness (ICW) in year t, and zero otherwise.
MA	=	Reputation indicator that equals one if the firm is on the Fortune’s America’s Most Admired Companies List (MA List) in year t, and zero otherwise.
Size	=	Firm size, calculated as the natural log of total assets (AT) at the end of year t.
Age	=	Firm age, calculated as the natural log of years the firm has CCM data as of the end of year t.
Agg_Loss	=	Loss indicator that equals one if earnings before extraordinary items (IB) in year t and year t − 1 sum to less than zero, and zero otherwise.
Extreme_Growth	=	Growth indicator that equals one if year-over-year industry adjusted (industry defined based on the 2-digit SIC code) sales (SALE) growth in year t falls into the top quintile, and zero otherwise.
ROA	=	Return on assets, calculated as earnings before extraordinary items (IB) in year t scaled by lagged total assets (AT).
D_XFIN	=	Newly issued debt and equity, calculated as the sum of cash received from the sale of stock and issuance of long-term debt, minus cash used in repurchase of stock, payment of dividends, and reduction in debt (SSTK + DLTIS − PRSTKC − DV − DLTR + DLCCH) in year t, scaled by average total assets (AT).
Agg_M&A	=	M&A indicator that equals one if the firm has non-zero acquisition expense (AQP) in years t and t − 1, and zero otherwise.
Foreign_Trans	=	Foreign transaction indicator that equals one if the firm has a non-zero foreign currency translation (FCA) in year t, and zero otherwise.
Segments	=	Number of segments, calculated as the natural log of the sum of the number of operating and geographic segments in year t.
Big4	=	Big 4 indicator that equals one if the firm engaged one of the largest four audit firms in year t. The largest four audit firms include Deloitte, Ernst&Young, KPMG, and PricewaterhouseCoopers.
Restatement	=	Restatement indicator that equals one if the firm has restatement in year t.
Auditor_Change	=	Auditor change indicator that equals one if the firm experiences an auditor change in year t.
Audit_Fee	=	Audit fees, calculated as audit fees scaled by the square root of total assets (AT) in year t.
High_ROA	=	Performance indicator that equals one if the firm has above-sample-median ROA, and zero otherwise.
Office_Size	=	Auditor office size indicator that equals one if the auditor office has above-sample-median total audit fees, and zero otherwise.

.

3.4. Sample

The sample period covers the years 2014–2022. It starts in 2014 because the earliest MA list available to us is from 2014.3 It ends in 2022 due to the availability of financial data as of the time we conducted this research. We obtain the internal control material weaknesses data from Audit Analytics. To construct our sample for primary analysis, we start from the client year observations at the intersection between CRSP/Compustat Merged (CCM) Database and Audit Analytics 404 (b) weaknesses, which generates 29,782 observations. We exclude observations with missing values in control variables and observations in the financial industry (SIC code: 6000-6999). Our final sample for primary analysis contains 20,842 client year observations.

Table 2. Sample selection and characteristics.

Panel A. Sample Selection
Sample Selection			Observations
Firm year observations at the intersection between CCM Database and Audit Analytics SOX 404 Internal Controls with opinions issued by auditors in the period of years 2014–2022			29,782
Minus: Observations with missing values in control variables			(2250)
Minus: Observations in the financial industry (SIC code: 6000-6999)			(6690)
Final sample			20,842
Panel B. Sample Distribution by Fiscal Year
Fiscal Year	Frequency	Percentage	Cumulative Percentage
2014	2451	11.76	11.76
2015	2467	11.84	23.60
2016	2468	11.84	35.44
2017	2370	11.37	46.81
2018	2326	11.16	57.97
2019	2281	10.94	68.91
2020	2089	10.02	78.94
2021	2146	10.30	89.23
2022	2244	10.77	100.00
Total	20,842	100.00

Panel A exhibits this sample selection process. Table 2 Panel B describes the distribution of our sample across fiscal years. Client firms are distributed across years almost evenly in our sample from 2014 to 2022.

4. Results

4.1. Descriptive Statistics

In this section, we describe the summary statistics of variables and report the Pearson correlation matrix.

Table 3. Summary statistics.

Variable	N	Mean	S.D.	P25	P50	P75
ICW	20,842	0.057	0.232	0.000	0.000	0.000
MA	20,842	0.080	0.271	0.000	0.000	0.000
Size	20,842	7.787	1.786	6.528	7.693	8.950
Age	20,842	3.033	0.744	2.485	3.091	3.555
Agg_Loss	20,842	0.293	0.455	0.000	0.000	1.000
Extreme_Growth	20,842	0.178	0.383	0.000	0.000	0.000
ROA	20,842	0.011	0.154	−0.014	0.036	0.081
D_XFIN	20,842	0.011	0.165	−0.063	−0.017	0.028
Agg_M&A	20,842	0.507	0.500	0.000	1.000	1.000
Foreign_Trans	20,842	0.547	0.498	0.000	1.000	1.000
Segments	20,842	2.452	0.793	1.792	2.565	3.045
Big4	20,842	0.852	0.356	1.000	1.000	1.000
Restatement	20,842	0.063	0.243	0.000	0.000	0.000
Auditor_Change	20,842	0.039	0.195	0.000	0.000	0.000
Audit_Fee	20,842	50.920	29.150	28.730	45.700	67.030

Notes: This table shows the summary statistics, including the number of observations, mean, standard deviation, the bottom quartile, the median, and the top quartile, of all variables in our primary analysis. Variable definitions are reported in Table 1.

presents the summary statistics of the variables in our main sample. Consistent with the frequency of internal control material weaknesses reported in prior research (Y. Chen et al., 2016, 2017), an average of 5.7 percent client year observations in our sample report SOX 404 (b) internal control material weaknesses. An average of 8.0 percent of client year observations in our sample appear on the MA list.

The means of Size, Age, ROA, D_XFIN, Segments, and Audit_Fee in our sample are 7.787, 3.0334, 0.011, 0.011, 2.452, and 50.920, respectively, consistent with the features of Compustat firms on average. An average of 29.3 percent of client year observations in our sample report losses in the recent consecutive two years (Agg_Loss), 17.8 percent of client year observations are exposed to extreme growth in the current year (Extreme_Growth), 50.7 percent of client year observations have M&A activities in the recent consecutive two years (Agg_M&A), 85.2 percent of client year observations employ Big 4 auditors to conduct audits in the current year (Big4), 6.3 percent of client year observations involve a restatement in the current year (Restatement), and 3.9 percent of client year observations experience an auditor switch in the current year (Auditor_Change).

Table 4. Pearson’s correlation matrix.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)
(1) ICW	1.0
(2) MA	−0.1 *	1.0
	(0.0)
(3) Size	−0.1 *	0.4 *	1.0
	(0.0)	(0.0)
(4) Age	−0.1 *	0.2 *	0.3 *	1.0
	(0.0)	(0.0)	(0.0)
(5) Agg_Loss	0.1 *	−0.2 *	−0.3 *	−0.3 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)
(6) Extreme_Growth	0.0	−0.0	0.0 *	−0.0 *	−0.0 *	1.0
	(0.5)	(0.3)	(0.0)	(0.0)	(0.0)
(7) ROA	−0.1 *	0.1 *	0.3 *	0.2 *	−0.6 *	0.1 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)
(8) D_XFIN	0.0 *	−0.1 *	−0.2 *	−0.2 *	0.3 *	0.1 *	−0.5 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)
(9) Agg_M&A	0.0 *	0.0	0.1 *	0.0	−0.1 *	0.0 *	0.1 *	0.0 *	1.0
	(0.0)	(0.1)	(0.0)	(0.8)	(0.0)	(0.0)	(0.0)	(0.0)
(10) Foreign_Trans	−0.0 *	0.0 *	−0.1 *	0.0 *	−0.0 *	0.0 *	−0.0 *	0.0 *	−0.0 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)
(11) Segments	−0.0	0.1 *	0.3 *	0.2 *	−0.2 *	−0.1 *	0.2 *	−0.1 *	0.1 *	−0.3 *	1.0
	(0.3)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)
(12) Big4	−0.1 *	0.1 *	0.4 *	0.0 *	−0.1 *	−0.0 *	0.1 *	−0.1 *	0.0 *	−0.1 *	0.1 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)
(13) Restatement	0.0 *	−0.0	−0.1 *	0.0 *	−0.0	−0.0 *	−0.0	0.0	−0.0	0.0 *	0.1 *	−0.1 *	1.0
	(0.0)	(0.3)	(0.0)	(0.0)	(0.8)	(0.0)	(0.0)	(0.0)	(0.3)	(0.0)	(0.0)	(0.0)
(14) Audtor_Change	0.1 *	−0.0 *	−0.1 *	−0.0 *	0.0 *	0.0	−0.0 *	0.0 *	−0.0	−0.0	−0.0	−0.1 *	0.0 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.2)	(0.3)	(0.3)	(0.0)	(0.0)
(15) Audit_Fee	0.2 *	0.1 *	−0.1 *	0.0 *	0.1 *	−0.1 *	−0.2 *	0.0	0.2 *	−0.1 *	0.2 *	0.1 *	−0.0	−0.1 *	1.0
	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.0)	(0.4)	(0.0)

Notes: * denotes that the Pearson correlation coefficient is significant at the 10% level. p-values of the Pearson correlation coefficients are reported in parentheses. Variable definitions are reported in Table 1.

provides the Pearson correlation matrix. The internal control material weakness indicator (ICW) and the MA indicator (MA) are significantly and negatively correlated. The Pearson correlation coefficient between the two indicators is −0.051.

4.2. Primary Results

In this section, we conduct the primary regression analysis for the association between ICW and MA using the model specified in Equation (1). The results are presented in

Table 5. Corporate reputation and internal control material weakness.

	(1)	(2)
VARIABLES	ICW	ICW
MA	−0.566 ***	−0.334 ***
	(−6.74)	(−3.54)
Size		−0.035 **
		(−2.07)
Age		−0.134 ***
		(−4.50)
Agg_Loss		0.158 ***
		(3.28)
Extreme_Growth		0.044
		(0.95)
ROA		−0.356 **
		(−2.44)
D_XFIN		−0.029
		(−0.27)
Agg_M&A		0.072 *
		(1.82)
Foreign_Trans		−0.048
		(−1.11)
Segments		0.023
		(0.68)
Big4		−0.405 ***
		(−7.17)
Restatement		0.255 ***
		(4.63)
Auditor_Change		0.533 ***
		(9.03)
Audit_Fee		0.010 ***
		(13.64)
Constant	−1.909 ***	−1.610 ***
	(−5.16)	(−4.28)
Fixed Effects	Year, Industry	Year, Industry
Observations	20,450	20,450
Pseudo R2	0.0280	0.116

Notes: This table shows the Probit estimation results for the association between internal control quality and company reputation. The dependent variable is ICW, an indicator for internal control material weaknesses of the client firm in a given year. The variable of interest is MA, an indicator for appearance on the MA list in a given year. Column (1) includes the variable of interests, year fixed effects, and industry fixed effects. Column (2) includes all variables. We cluster the standard error by client firms. All variable definitions are shown in Table 1. *, **, and *** indicate significance levels at less than 10%, 5%, and 1% based on two-tailed t-tests.

.5

Column (1) shows the baseline results with only the key variable of interests, MA, and fixed effects included on the right-hand side of the regression equation. The results in column (1) show that the coefficient of the MA indicator (MA) is negative and highly significant, suggesting that high-reputation client firms (those on the MA list) have a lower probability of internal control material weakness, and thus have higher internal control quality. The results are consistent with our H1. Column (2) exhibits the results of the full regression model. After controlling variables that are correlated with the probability of having internal control material weaknesses, the coefficient of MA remains negative and highly significant, further supporting our H1. Based on our back-of-the-envelope calculation, the coefficient of MA (= −0.334) in column (2) suggests that the probability of ICW decreases by approximately 13.1 percent when MA changes from 0 to 1. This economic magnitude is large enough compared with existing determinants of internal control material weaknesses, such as ROA. Overall, our results suggest a negative association between company reputation and internal control material weakness.

Our results align with the findings of Huang and Kang (2018, 2022) who demonstrate that high-reputation firms have stronger incentives to engage high-quality auditors and to maximize the potential benefits of auditors’ knowledge spillover. Similarly, we find that high-reputation firms have strong incentives to protect their reputation by minimizing the internal control material weaknesses disclosed by auditors. These findings contribute to the audit literature by highlighting the potential consequences of corporate reputation for auditor-related outcomes.

Our results also corroborate those of Cao et al. (2012), who show that high-reputation firms maintain a superior financial reporting quality and establish a link between corporate reputation and corporate disclosures. We extend this line of research by demonstrating the impact of corporate reputation on disclosure quality beyond earnings quality, as identified by Cao et al. (2012), and by establishing a link between corporate reputation and internal control quality.

4.3. Cross-Sectional Analysis Results

Next, we conduct our cross-sectional analyses to examine whether the client firm’s incentive to protect its reputation is the mechanism through which reputable companies are less likely to have internal control material weaknesses. We consider such incentives from client companies’ pressure to maintain a good reputation and auditors’ quality of monitoring clients to uphold this reputation.

We capture client companies’ pressure to maintain a good reputation with firm performance. Signaling theory suggests that a company’s reputation can serve as a signal to various stakeholders, with better-quality companies (i.e., better-performing companies) having stronger incentives to invest in this signal to separate themselves from lower-quality companies (Akerlof, 1970; Bergh et al., 2010; Spence, 1973, 1974). We use ROA, a financial performance measure, to construct our grouping variable for better-performing clients, referred to as High_ROA. We conduct the regression analysis for this cross-sectional test using Equation (2).

$$\begin{array}{cc}\hfill IC{W}_{i,t}= {\beta }_0& + {\beta }_{1 }Hig{h\text{\_}ROA}_{i,t }+{\beta }_2{MA}_{\mathrm{i},\mathrm{t}}+{\beta }_{3 }{MA}_{\mathrm{i},\mathrm{t}}\times Hig{h\text{\_}ROA}_{i,t }+{\beta }_4 {Size}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{5 }{Age}_{\mathrm{i},\mathrm{t}}+ {\beta }_6 {Agg\text{\_}Loss}_{\mathrm{i},\mathrm{t}}+{\beta }_7 {Extreme\text{\_}Growth}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_8 {ROA}_{\mathrm{i},\mathrm{t}}+{\beta }_9 {D\text{\_}XFIN}_{\mathrm{i},\mathrm{t}}+{\beta }_{10 }{Agg\text{\_}M\&A}_{\mathrm{i},\mathrm{t}}+{\beta }_{11}{ Foreign\text{\_}Trans}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{12} {Segments}_{\mathrm{i},\mathrm{t}}+{\beta }_{13} {Big4}_{\mathrm{i},\mathrm{t}}+{\beta }_{14} {Restatement}_{\mathrm{i},\mathrm{t}}+{\beta }_{15} {Auditor\text{\_}Change}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{16} {Audit\text{\_}Fee}_{\mathrm{i},\mathrm{t}}+Year Fixed Effects+Industry Fixed Effects+u\hfill \end{array}$$

(2)

where i denotes client firm i and t denotes year t. We estimate Equation (2) using a Probit model. High_ROA is a performance indicator equal to one if the firm has an above-median ROA, and zero otherwise. The coefficient of the interaction term, ${\beta }_3$, is the coefficient of interest in this cross-sectional analysis and is predicted to be negative if better-performing reputable clients have lower probability of internal control material weakness. We present the results of the first cross-sectional tests in

Table 6. Cross-sectional analysis results: client performance.

	(1)	(2)
VARIABLES	ICW	ICW
High_ROA	−0.470 ***	−0.313 ***
	(−12.48)	(−6.55)
MA	−0.573 ***	−0.381 **
	(−4.11)	(−2.57)
MA × High_ROA	−0.846 ***	−0.546 ***
	(−8.39)	(−4.83)
Size		−0.042 **
		(−2.50)
Age		−0.130 ***
		(−4.34)
Agg_Loss		0.037
		(0.75)
Extreme_Growth		0.065
		(1.42)
ROA		−0.027
		(−0.17)
D_XFIN		0.012
		(0.11)
Agg_M&A		0.058
		(1.46)
Foreign_Trans		−0.042
		(−0.96)
Segments		0.024
		(0.70)
Big4		−0.405 ***
		(−7.20)
Restatement		0.254 ***
		(4.59)
Auditor_Change		0.528 ***
		(8.95)
Audit_Fee		0.009 ***
		(13.51)
Constant	−1.811 ***	−1.461 ***
	(−4.96)	(−3.96)
Fixed Effects	Year, Industry	Year, Industry
Observations	20,450	20,450
Pseudo R2	0.0533	0.122

Notes: This table reports the moderating effects of client performance. The dependent variable is ICW, an indicator for internal control material weaknesses of the client firm. MA is an indicator for appearance on the MA list. High_ROA is an indicator for better-performing clients, equal to one if the client firm has an above-median ROA, and zero otherwise. The variable of interest is the interaction term, MA × High_ROA. Column (1) includes the main effects, the interaction term, year fixed effects, and industry fixed effects. Column (2) includes all variables. We cluster the standard error by client firms. All variable definitions are shown in Table 1. *, **, and *** indicate significance levels at less than 10%, 5%, and 1% based on two-tailed t-tests.

.

Column (1) shows the baseline results with only the main effects, MA and High_ROA, and the interaction term, MA × High_ROA, as well as fixed effects included on the right-hand side of the regression equation. Consistent with our prediction for the first cross-sectional analysis, the results in column (1) show that the coefficient of the interaction (MA × High_ROA) is negative and highly significant, suggesting that better-performing reputable client firms have a lower probability of internal control material weakness, and thus have higher internal control quality. Column (2) exhibits the results of the full regression model. After controlling variables that are correlated with the probability of internal control material weaknesses, the coefficient of the interaction term remains negative and highly significant, further supporting our prediction for the first cross-sectional analysis.

Next, we move on to the second cross-sectional analysis, which examines the auditor’s quality of monitoring clients to uphold a good reputation. Prior studies have shown that larger auditor offices provide better audit quality (Choi et al., 2010; Francis & Yu, 2009) and stronger monitoring effects on clients’ decisions (Whitworth & Lambert, 2014). This indicates that larger auditor offices can monitor clients more effectively to maintain a good reputation. We construct a grouping variable, Office_Size, to capture large auditor office following the existing literature (Choi et al., 2010; Francis & Yu, 2009; Goldie et al., 2018). We conduct the regression analysis for this cross-sectional test using Equation (3).

$$\begin{array}{cc}\hfill IC{W}_{i,t}= {\beta }_0& + {\beta }_{1 }{Office\text{\_}Size}_{i,t }+{\beta }_2{MA}_{\mathrm{i},\mathrm{t}}+{\beta }_{3 }{MA}_{\mathrm{i},\mathrm{t}}\times {Office\text{\_}Size}_{i,t }+{\beta }_4 {Size}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{5 }{Age}_{\mathrm{i},\mathrm{t}}+ {\beta }_6 {Agg\text{\_}Loss}_{\mathrm{i},\mathrm{t}}+{\beta }_7 {Extreme\text{\_}Growth}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_8 {ROA}_{\mathrm{i},\mathrm{t}}+{\beta }_9 {D\text{\_}XFIN}_{\mathrm{i},\mathrm{t}}+{\beta }_{10 }{Agg\text{\_}M\&A}_{\mathrm{i},\mathrm{t}}+{\beta }_{11}{ Foreign\text{\_}Trans}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{12} {Segments}_{\mathrm{i},\mathrm{t}}+{\beta }_{13} {Big4}_{\mathrm{i},\mathrm{t}}+{\beta }_{14} {Restatement}_{\mathrm{i},\mathrm{t}}+{\beta }_{15} {Auditor\text{\_}Change}_{\mathrm{i},\mathrm{t}}\hfill \\ & + {\beta }_{16} {Audit\text{\_}Fee}_{\mathrm{i},\mathrm{t}}+Year Fixed Effects+Industry Fixed Effects+u\hfill \end{array}$$

(3)

where i denotes client firm i and t denotes year t. We estimate Equation (3) using a Probit model. Office_Size is an indicator equal to one if the auditor office has above-sample-median total audit fees, and zero otherwise. The coefficient of the interaction term, ${\beta }_3$, is the coefficient of interest in this cross-sectional analysis and is predicted to be negative if reputable clients audited by large auditor office have a lower probability of internal control material weakness. We present the results of the second cross-sectional tests in

Table 7. Cross-sectional analysis results: auditor office size.

	(1)	(2)
VARIABLES	ICW	ICW
Office_Size	−0.165 ***	−0.182 ***
	(−3.59)	(−3.50)
MA	−0.607 ***	−0.387 **
	(−4.02)	(−2.37)
MA × Office_Size	−0.660 ***	−0.450 ***
	(−6.51)	(−3.81)
Size		−0.042 **
		(−2.09)
Age		−0.099 ***
		(−3.01)
Agg_Loss		0.107 **
		(1.96)
Extreme_Growth		0.096 *
		(1.88)
ROA		−0.344 **
		(−2.11)
D_XFIN		−0.015
		(−0.12)
Agg_M&A		0.127 ***
		(2.94)
Foreign_Trans		−0.029
		(−0.60)
Segments		−0.004
		(−0.09)
Big4		−0.359 ***
		(−5.25)
Restatement		0.302 ***
		(5.20)
Auditor_Change		0.486 ***
		(7.06)
Audit_Fee		0.011 ***
		(13.87)
Constant	−1.854 ***	−1.657 ***
	(−5.02)	(−4.37)
Fixed Effects	Year, Industry	Year, Industry
Observations	16,924	16,924
Pseudo R2	0.0355	0.130

Notes: This table reports the moderating effects of auditor office size. The dependent variable is ICW, an indicator for internal control material weaknesses of the client firm. MA is an indicator for appearance on the MA list. Office_Size is an indicator for large auditor office size, equal to one if the auditor office has above-median total audit fees, and zero otherwise. The variable of interest is the interaction term, MA × Office_Size. Column (1) includes the main effects, the interaction term, year fixed effects, and industry fixed effects. Column (2) includes all variables. We cluster the standard error by client firms. All variable definitions are shown in Table 1. *, **, and *** indicate significance levels at less than 10%, 5%, and 1% based on two-tailed t-tests.

.

Column (1) shows the baseline results with only the main effects, MA and Office_Size, and the interaction term, MA × Office_Size, as well as fixed effects included on the right-hand side of the regression equation. Consistent with our prediction for the second cross-sectional analysis, the results in column (1) show that the coefficient of the interaction (MA × Office_Size) is negative and highly significant, suggesting that reputable clients audited by large auditor offices have a lower probability of internal control material weakness, and thus have higher internal control quality. Column (2) exhibits the results of the full regression model. After controlling variables that are correlated with the probability of internal control material weaknesses, the coefficient of the interaction term remains negative and highly significant, further supporting our prediction for the second cross-sectional analysis.

Taken together, our cross-sectional analyses suggest that the client firm’s incentive to protect its reputation, driven by the need to signal a strong performance and by monitoring pressure from high-quality auditors, serves as the mechanism through which reputable clients are less likely to exhibit internal control material weaknesses.

4.4. Robustness Checks

In this section, we conduct a sequence of alternative analyses as robustness checks. First, for the key variable of interest, we replace the MA indicator in Equation (1) with a continuous variable, MA_Year. MA_Year captures the number of years the firm appears on the MA list and is calculated as the total number of sample years to date during which the firm has appeared on the MA list. The higher the value of MA_Year, the more frequently a client firm appears on the MA list, indicating a higher company reputation. The regression results are shown in

Table 8. Alternative reputation measure.

	(1)	(2)
VARIABLES	ICW	ICW
MA_Year	−0.140 ***	−0.086 ***
	(−5.36)	(−3.26)
Size		−0.034 **
		(−1.99)
Age		−0.130 ***
		(−4.37)
Agg_Loss		0.159 ***
		(3.29)
Extreme_Growth		0.045
		(0.99)
ROA		−0.362 **
		(−2.49)
D_XFIN		−0.033
		(−0.30)
Agg_M&A		0.071 *
		(1.80)
Foreign_Trans		−0.048
		(−1.11)
Segments		0.022
		(0.66)
Big4		−0.405 ***
		(−7.19)
Restatement		0.255 ***
		(4.64)
Auditor_Change		0.534 ***
		(9.04)
Audit_Fee		0.010 ***
		(13.66)
Constant	−1.944 ***	−1.652 ***
	(−5.26)	(−4.35)
Fixed Effects	Year, Industry	Year, Industry
Observations	20,450	20,450
Pseudo R2	0.0286	0.117

Notes: This table shows the Probit estimation results for the association between internal control quality and company reputation. The dependent variable is ICW, an indicator for internal control material weaknesses of the client firm in a given year. The variable of interest is MA_Year, the number of sample years to date during which the client firm appears on the MA list. Column (1) includes the variable of interests, year fixed effects, and industry fixed effects. Column (2) includes all variables. We cluster the standard error by client firms. All variable definitions are shown in Table 1. *, **, and *** indicate significance levels at less than 10%, 5%, and 1% based on two-tailed t-tests.

. Similarly to our primary results, the coefficient on MA_Year remains negative and highly significant, suggesting that the probability of internal control material weaknesses is lower when a client firm has a higher company reputation. The results of this alternative analysis support our main hypothesis, H1, which predicts a negative association between company reputation and internal control material weaknesses.

Next, rather than using the ICW indicator and conducting the Probit estimation, we replace the ICW indicator in Equation (1) with a count variable, ICW_Count, and run the OLS regression for Equation (1). ICW_Count is defined as the total number of SOX 404 (b) internal control material weaknesses identified by auditors. The higher the value of ICW_Count, the more internal control material weaknesses that a client firm has. We report the OLS regression results in

Table 9. Alternative ICW measure.

	(1)	(2)
VARIABLES	ICW_Count	ICW_Count
MA	−1.349 ***	−0.869 ***
	(−5.50)	(−3.39)
Size		−0.065 *
		(−1.76)
Age		−0.188 ***
		(−2.80)
Agg_Loss		0.438 ***
		(3.95)
Extreme_Growth		0.097
		(1.01)
ROA		−0.215
		(−0.73)
D_XFIN		0.238
		(1.01)
Agg_M&A		0.190 **
		(2.17)
Foreign_Trans		−0.016
		(−0.17)
Segments		0.069
		(0.84)
Big4		−0.637 ***
		(−5.21)
Restatement		0.546 ***
		(5.04)
Auditor_Change		0.990 ***
		(9.26)
Audit_Fee		0.018 ***
		(14.44)
Constant	−2.944 ***	−2.839 ***
	(−3.40)	(−3.31)
Fixed Effects	Year, Industry	Year, Industry
Observations	20,842	20,842
Pseudo R2	0.0452	0.153

Notes: This table shows the OLS regression results for the association between internal control quality and company reputation. The dependent variable is ICW_Count, the total number of SOX 404 (b) internal control material weaknesses of the client firm in a given year. The variable of interest is MA, an indicator for appearance on the MA list in a given year. Column (1) includes the variable of interests, year fixed effects, and industry fixed effects. Column (2) includes all variables. We cluster the standard error by client firms. All variable definitions are shown in Table 1. *, **, and *** indicate significance levels at less than 10%, 5%, and 1% based on two-tailed t-tests.

. Consistent with our primary results, the coefficient on MA remains negative and highly significant, suggesting that the number of internal control material weaknesses is lower when a client firm appears on the MA list. This supports our main hypothesis of a negative association between internal control material weaknesses and company reputation.

Lastly, a potential concern for our main results is the endogenous nature of our key variable of interest, MA. It is likely that some unobservable characteristics correlated with both corporate reputation and internal control material weakness are omitted in our regression. To address this endogeneity concern, we conduct two-stage least squares (2SLS) regressions with the Inverse Mills Ratio (i.e., the Heckman two-stage selection model) to re-examine our hypothesis, following Cao et al. (2012) and Huang and Kang (2018). Specifically, in the first stage, we run a Probit regression to estimate the probability of being selected for the MA list on three instrumental variables identified by Cao et al. (2012) and Huang and Kang (2018): research and development (R&D) intensity, advertising intensity, and the number of employees. We then calculate the Inverse Mills Ratios (IMR) from the first-stage regression. In the second stage, we run OLS regressions using the IMR estimated from the first stage along with all controls in Equation (1), restricting the sample to selected observations (i.e., those with MA = 1). We present the second-stage results in

Table 10. The second-stage results of Heckman’s two-stage selection model.

	(1)	(2)
VARIABLES	ICW	ICW
IMR	0.000	−0.008
	(0.01)	(−0.60)
Size		−0.010 *
		(−1.73)
Age		0.004
		(0.55)
Agg_Loss		0.052 *
		(1.68)
Extreme_Growth		−0.005
		(−0.43)
ROA		−0.039
		(−0.38)
D_XFIN		0.061
		(1.00)
Agg_M&A		0.009
		(1.53)
Foreign_Transaction		0.004
		(0.38)
Segments		−0.020 **
		(−2.03)
Big4		0.064 ***
		(2.94)
Restatement		0.010
		(0.60)
Auditor_Change		−0.008
		(−0.74)
Audit_Fee		0.001 ***
		(3.13)
Constant	−0.005	0.033
	(−0.31)	(0.36)
Fixed Effects	Year, Industry	Year, Industry
Observations	1665	1665
R-squared	0.037	0.075

Notes: This table presents the second-stage results of the Heckman selection model. The dependent variable is ICW, an indicator for internal control material weaknesses of the client firm. IMR is the Inverse Mills Ratio (IMR) obtained from the first-stage Probit regression, in which we regress our key variable of interest, MA, on three instrumental variables: research and development (R&D) intensity, advertising intensity, and the number of employees. Column (1) includes IMR, year fixed effects, and industry fixed effects. Column (2) includes all variables. We cluster the standard error by client firms. All variable definitions except for IMR are shown in Table 1. *, **, and *** indicate significance levels at less than 10%, 5%, and 1% based on two-tailed t-tests.

. Consistent with Cao et al. (2012) and Huang and Kang (2018), we find that the coefficient of IMR is insignificant, suggesting that self-selection is not a significant concern in our model.

5. Conclusions

Our study investigates the association between company reputation and internal control quality. Using data from the Fortune 1000 list for the years 2014 to 2022, we find that high-reputation companies have a lower probability of internal control material weaknesses, suggesting lower internal control risk and higher internal control quality compared with their counterparts. These findings are more pronounced for high-ROA client firms and client firms audited by large auditor offices, suggesting that reputable companies with better financial performances or those audited by larger office sizes show a further reduced likelihood of internal control material weaknesses, reinforcing their higher internal control quality. Additionally, our results hold when using alternative measures: the number of years a company appears on the MA list as a proxy for company reputation, and the total number of SOX 404 (b) internal control material weaknesses identified by auditors as a measure of internal control quality. These alternative analyses provide additional supporting evidence that companies with higher reputations are less likely to experience internal control material weaknesses, indicating a stronger quality of internal controls.

In summary, our study demonstrates that higher-reputation companies are motivated to protect their reputation, driven by the need to signal a strong performance and by the monitoring pressure from high-quality auditors. As a result, these high-reputation companies are less likely to exhibit internal control material weaknesses, which reflects lower internal control risk and stronger internal control quality. This study contributes to the expanding literature on the impact of corporate reputation on firm behavior and decision-making processes and to the existing literature on the factors influencing internal control quality.

Despite these contributions, our study has limitations that offer opportunities for future research. Our sample focuses on large U.S. firms on the Fortune 1000 list, which may limit the generalizability of our findings. Future research could examine whether the documented association holds in other contexts or among smaller firms. Additionally, researchers could explore whether this association varies across different time periods or in response to major economic events.