Sage Journals: Discover world-class research

Abstract

Background

While whistleblowing (WB) has attracted growing research interest in recent years, several critical WB-related issues remain underexplored.

Purpose

This study examines the impact of external WB allegations on a firm’s organizational capital (OC). Such allegations often indicate management’s failure to address employee concerns internally, spotlighting potential deficiencies in internal reporting systems, employee communication, training, and trust in organizational fairness. To mitigate reputational damage, restore employee trust, and prevent future incidents, we posit that WB firms respond by increasing OC investment.

Research Design

We employ a difference-in-differences approach, comparing OC changes in WB-targeted firms with those in a propensity score-matched control sample.

Study Sample

Our dataset includes employee WB allegations obtained from OSHA (via a Freedom of Information Act request) and a hand-collected sample from public media.

Results

We find that WB firms significantly increase OC in the post-allegation period. Additionally, higher OC investment is linked to fewer future WB incidents. The decision to strengthen OC is primarily influenced by employees, long-term institutional investors, and prior OC deficiencies, rather than by WB case credibility or CEO characteristics.

Conclusions

Our findings indicate the importance of aligning long-term investment strategies and employee benefits with broader corporate goals to foster a responsive and adaptive organizational culture.

Keywords

whistleblowing financial reporting misconduct organizational capital

Introduction

Employee whistleblowing (WB) refers to “the disclosure by organization members of illegal, immoral, or illegitimate practices under control of their employers, to persons or organizations that may be able to effect action” (Near & Miceli, 1985, p. 4). Whistleblowers are key to helping firms detect fraud and organizational deficiencies, because they are usually among the first to observe wrongdoing (Brief & Motowidlo, 1986; Dyck et al., 2010; Lee & Fargher, 2013). Employee whistleblowers typically prefer to report concerns internally. This is because external WB is associated with higher costs of retaliation, discrimination, and unemployment. Thus, external WB allegations often reflect management’s failure to address concerns internally, due to inadequate reporting systems, unclear policies, or a lack of employee trust in organizational fairness and justice (Lee & Fargher, 2013; Stubben & Welch, 2020). Once WB allegations surface publicly, the WB target firms (WB firms or target firms, hereafter) face greater scrutiny from the public and regulators, significant reputational penalties, and escalating litigation risk (Bowen et al., 2010).

This study extends the WB literature by examining the relation between external WB and target firms’ organizational capital (OC). OC comprises the cumulative knowledge and collection of business processes and systems that enable firms to operate more efficiently. It includes investment in workforce training, employee voice, communications, and work design (Black & Lynch, 2005; Lev et al., 2009; Li et al., 2018). We argue that external WB allegations likely reflect deficient internal reporting systems, unclear policies and communications, and underinvestment in employees to enhance their knowledge, capabilities, and trust in organizational fairness and justice. In other words, external WB allegations reflect inadequate investment in OC. Therefore, we expect target firms to respond to allegations by increasing investment in OC.

Building on prior research (e.g., Bowen et al., 2010; Kuang et al., 2021), we use two data sources to identify external WB allegations. First, we filed a Freedom of Information Act (FOIA) request with the Occupational Safety and Health Administration (OSHA) (under the U.S. Department of Labor) to obtain WB data. Under §806 of the Sarbanes-Oxley Act of 2002 (SOX), employees can file discrimination complaints with OSHA that stem from alleged financial misconduct. OSHA is then required to communicate these allegations to the U.S. Securities and Exchange Commission (U.S. Department of Labor, 2012). Second, we supplement the OSHA sample with a hand-collected sample of media-reported WB allegations.

Following Eisfeldt and Papanikolaou (2013) and Li et al. (2018), we measure OC as capitalized selling, general, and administrative (SG&A) expenses divided by total assets. To address any endogeneity concerns related to the non-random nature of WB allegations, we create a control sample of non-WB firms using a propensity score matching (PSM) procedure (PSM control sample).¹ We then use a difference-in-differences (DiD) research design to compare changes in WB targets’ OC with that of the PSM control sample.

Our univariate tests and multivariate regression results show that WB firms respond to allegations by increasing OC. The increase is statistically and economically significant. For an average WB firm in our sample, the change represents a $45.74 million increase in OC. To corroborate our PSM sample findings, we also conduct a robustness test using an entropy-balanced sample, and find that our inferences hold.

Concerns about using SG&A expenses to measure OC may arise due to potential measurement errors from unrelated costs, such as advertising, R&D, and legal expenses. These may introduce noise. To address this issue, we conduct multiple robustness checks using alternative OC measures. Following Li et al. (2018) and Danielova et al. (2023), we use the industry median-adjusted ratio of OC to total assets and the ratio of SG&A expenses to total assets. Additionally, we subtract legal expenses from SG&A, and recalculate OC in order to mitigate concerns that higher OC may reflect litigation-related expenses. We further remove litigation, advertising, and R&D expenses from SG&A, and reestimate our tests. The results confirm that WB firms increase OC investment post-allegation.

Another concern is that OC may not directly measure the costs of improving internal communication to prevent WB. Investments in WB hotlines, training, and corporate citizenship programs are rarely disclosed. To overcome this measurement challenge, we use qualitative measures of OC that are closely aligned with our underlying constructs. Specifically, we obtain four qualitative OC dimensions from the Refinitiv ESG database that capture a company’s WB communication tools and policies: business ethics, employee involvement, employee training, and anti-fraud activities. These measures reflect aspects of OC related to reinforcing employee knowledge and trust, improving corporate citizenship, and enhancing companywide policies against bribery, corruption, and fraud.

Our robustness checks indicate a strong correlation between the original measure of OC and each of the four qualitative dimensions, as well as the composite measure from their sum. Our main inferences remain unchanged when using (1) the composite qualitative OC measure, (2) each individual qualitative dimension, or (3) the expected value of OC estimated by regressing the composite qualitative OC on the original OC while controlling for fixed effects.

Overall, our results from the robustness tests using alternative measures align with our main results. External WB allegations prompt target firms to increase investment in OC by enhancing WB communications and policies related to business ethics, employee involvement, training, and anti-fraud activities. This, in turn, promotes good corporate citizenship and greater awareness of anti-fraud policies.

Next, given that 75.16% of WB firms in our sample increased OC post-allegation, we use logistic regressions to explore which factors influenced this action. The results suggest that labor intensity, negative market reactions to an allegation, lower levels of prior OC, and long-term institutional investors are the primary drivers of the decision to increase OC. In contrast, factors such as WB case merit, short-term institutional investors, and CEO factors like ownership, risk-taking incentives, gender, or duality did not affect the decision.

Thus, our findings demonstrate that WB firms that are more heavily reliant on labor are more responsive to employee concerns. Furthermore, WB firms with low levels of pre-allegation OC are motivated to adjust their inadequate investment in OC post-allegation. Finally, our results are in line with those of prior studies documenting the influential monitoring role of long-term institutional investors (e.g., Attig et al., 2012, 2013). We find they appear to encourage WB firms to focus on long-term goals by increasing investment in OC post-allegation.

We also examine how changes in OC affect the likelihood of future external WB incidents. If post-allegation investments in OC improve internal reporting, organizational processes, and communication, fewer external WB cases should follow.² Our findings show that WB firms that increase OC investments experience fewer WB incidences in the two years post-allegation. They also have a lower probability of being external WB targets than their counterparts that did not increase OC. Untabulated analyses show that WB targets exhibit a post-allegation increase in OC regardless of whether the employee concerns were reported to the media or to OSHA.

Finally, we examine whether WB targets increase brand capital (i.e., capitalized advertising expenses) post-allegation. We aim to determine whether they are making cosmetic changes to mitigate reputational harm, or trying to seriously address the root causes of the allegations. Our results show that the average WB firm in our sample does not increase brand capital. This suggests that the documented increases in OC are intended to address the inadequacies that led to the WB allegation.

Our paper contributes to the WB and OC literature in several key ways. First, prior research on WB focuses primarily on determinants (Andrade, 2015; Cheng et al., 2019; Culiberg & Mihelič, 2017; Gao & Brink, 2017; Hartman & Desjardins, 2008; Hoffman & Schwartz, 2015; Morrison & Milliken, 2000). There is only a limited focus on target firm responses to WB allegations. Notable exceptions include Wilde (2017), who finds that WB firms reduce financial misreporting and tax aggressiveness, and Bowen et al. (2010), who show that WB firms subsequently improve corporate governance. Some earlier studies (e.g., Gobert & Punch, 2000; Miceli & Near, 1992; Schmidt, 2005) argue that firms often view allegations as frivolous, and therefore fail to enact meaningful changes. However, the literature remains largely silent on whether and how WB firms adjust OC investments to facilitate internal reporting, strengthen control, and amplify employee voice. Our paper addresses this gap.

Second, our study enriches the OC literature by providing novel evidence on how WB firms respond to allegations. We demonstrate that external WB allegations often reflect deficiencies in internal reporting systems and communication, as well as underinvestment in employees. Our findings show that WB firms rectify these deficiencies by increasing OC investment.

Understanding the Whistleblowing Process

Institutional Background

WB allegations may be reported internally or externally. The academic literature generally supports the idea that WB is an effective mechanism for uncovering internal control weaknesses (Kuang et al., 2021), misconduct, and unethical practices (Dyck et al., 2010; Lee & Xiao, 2018). Since the passage of the False Claims Act of 1863, U.S. legislators have implemented various laws to protect whistleblowers from retaliation, and provide financial incentives as motivation to report wrongdoing (e.g., the Sarbanes-Oxley Act of 2002 (SOX) and the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank) of 2010).³

Under SOX, public firms are required to establish appropriate systems to receive complaints from employees about questionable accounting or auditing matters (SOX §301). In addition, public firms are required to document in their annual reports how management maintains adequate internal control structures (SOX §404). Thus, the law requires WB firms to improve their internal reporting and monitoring systems.

However, despite this law, recent studies find that internal control systems are not always adequate (Soltes, 2020; Stubben & Welch, 2020). Some firms establish procedures simply for compliance (“window-dressing”), but do not invest adequate resources to ensure the controls function as intended. For example, legislation in the U.S. requires firms to offer employees a means to report accounting misconduct anonymously, but significant barriers to a smooth reporting process often remain. As Soltes (2020) notes, approximately one-fifth of firms have impediments to internal reporting of accounting irregularities, and about one-tenth fail to respond to internal allegations. In addition, rapidly growing firms tend to receive fewer internal WB reports per employee, with less information, and less follow-up by management (Stubben & Welch, 2020).

Whistleblowing Process

Employee perceptions of organizational justice, fairness, and transparency influence potential whistleblowers at various stages. In the first stage, employees observe the alleged misconduct or unethical practice. In the second stage, they consider the incentives and disincentives of reporting the misconduct, as well as past organizational responses. Here, explicit statements of a code of conduct, organizational support for ethical behavior, whistleblower protection, and financial rewards are incentives to report the misconduct internally. In contrast, perceptions that WB will not result in corrective actions, and may lead to potential discrimination or retaliation, are likely to influence employees to remain silent about the misconduct. Prior research shows that, if employees decide to report misconduct, they first make an internal report to supervisors or managers (Bjørkelo et al., 2011; Smith, 2010).

The third stage describes possible organizational responses once an internal report is filed. Acceptable responses include thorough investigations and corrective actions to stop the unethical activity. Unacceptable responses include ignoring the WB report, or subjecting whistleblowers to discrimination or retaliation at work.

The final stage encompasses the post-internal allegation period. Here, a favorable path would mean the whistleblower is satisfied with the internal outcome, and the process ends. An unfavorable path could mean the whistleblower feels demoralized and remains silent, or seeks resolution by reporting the misconduct externally to authorities or the media. If the WB process progresses along this path, financial and reputational costs to target firms can dramatically increase.

Prior Literature and Hypothesis Development

Whistleblowing has attracted significant attention from the public, regulators, and scholars in recent years. Prior research shows that WB is one of the most effective methods for exposing corporate malfeasance (Lee & Xiao, 2018). Dyck et al. (2010) find that employee whistleblowers are more effective than outside monitors at detecting corporate fraud. The Association of Certified Fraud Examiners (2020) reports that approximately 43% of detected fraud cases are identified by employee whistleblowers. Furthermore, recent studies show that WB allegations are taken seriously by regulators, law enforcement agencies, and WB target firms (Bowen et al., 2010; Call et al., 2018; Kuang et al., 2021; Stubben & Welch, 2020; Wilde, 2017).

Employees who observe misconduct typically prefer to report their concerns internally due to the higher costs associated with external WB, such as retaliation, discrimination, and unemployment. However, if internal reports do not receive adequate attention, or if management fails to take corrective action, employees may resort to external WB by going public or reporting to a government agency. Stubben and Welch (2020) analyze nearly 2 million internal WB reports, and find significant variation in the efficacy of the internal systems, which they attribute to managers’ differing views on the costs and benefits of these systems. Some firms implement WB systems solely to comply with SOX. But they fail to actively promote or use them. In other firms, managers worry that WB systems could be misused by underperforming employees to file frivolous complaints to gain legal protection as whistleblowers.

Soltes (2020) conducts a comprehensive field study on the functionality of internal WB systems. He finds significant obstacles to WB reporting, including disconnected phone lines, incorrect web redirects, and email bouncebacks in 20% of firms’ reporting systems. Approximately 10% of firms fail to respond promptly to reports. Importantly, merely implementing a WB reporting system does not necessarily fulfill the intended purpose of preventing fraud and misconduct. The system will remain ineffective if firms do not appropriately train and encourage employees to use it, or if employees do not trust management to address reported claims. Although there is some evidence on the functionality of internal WB systems, little is yet known about whether and how target firms respond to external WB allegations.

External WB allegations not only reflect management’s failure to internally address employees’ concerns, they also highlight possible deficiencies in internal reporting systems, communication to employees, and underinvestment in employee training. These issues are captured in the four components of organizational capital investment identified by Black and Lynch (2005): workforce training, employee voice, communication, and work design.

OC is generally defined as the cumulative knowledge and collection of business processes and systems that enable firms to utilize their input resources more efficiently (Black & Lynch, 2005; Lev et al., 2009; Li et al., 2018). Firms develop this intangible asset through investment in their workforce and their operational practices and systems. Underinvestment in OC can disrupt operating processes and communication in a multitude of ways (Atkeson & Kehoe, 2005; Eisfeldt & Papanikolaou, 2013; Lev et al., 2009; Lev & Radhakrishnan, 2005; Prescott & Visscher, 1980).

Prosocial organizational behavior theory suggests that the actions firms take to address employee concerns are crucial to restoring and building an organization-wide perception of integrity, fairness, and justice (Miceli et al., 2008). Trevino and Weaver (2001) and Rupp and Bell (2010) provide evidence that employees’ perceptions of justice influence their willingness to go beyond their job obligations to advocate for justice. This behavior is essential for maintaining a positive firm culture, promoting the safe reporting of unethical practices and misconduct, and motivating employees to identify problems that could negatively impact organizational goals. The likelihood of internal WB thus increases when employees perceive fairness in organizational WB procedures and outcomes, and in their interactions with supervisors (Seifert et al., 2010).

Establishing effective communication channels for employees to report problems without fear of discrimination, repercussions, or retaliation is essential for fostering trust and transparency within an organization. Effective internal reporting of misconduct also enables management to promptly address and resolve employee concerns, as well as avoid reputational damage and lengthy, costly investigations. Therefore, when an employee bypasses the internal WB system to report corporate misconduct externally, management should feel compelled to address internal deficiencies and underinvestment in OC. Hence, we conjecture that firms will respond to external WB allegations by increasing investment in OC:

Target firms increase investment in organizational capital in the post-allegation period.

Research Design

Measuring Organizational Capital

We follow prior literature (Eisfeldt & Papanikolaou, 2013; Lev et al., 2016; Lev & Radhakrishnan, 2005; Li et al., 2018), and measure a firm’s stock of OC by using capitalized SG&A expenses divided by total assets. SG&A expenses include communication expenditures (e.g., information technology, information systems, and distribution channels), consulting fees, investment in workforce training costs, work design, business processes, and systems, advertising and marketing expenses, and employee salaries, benefits, and promotions. These expenses aim to improve a firm’s body of knowledge and the synergy between combinations of labor and capital investments in OC (Eisfeldt & Papanikolaou, 2013; Li et al., 2018). We compute the stock of OC by using the perpetual inventory method. We recursively add the deflated value of SG&A expenses as follows:

O C_{i, t} = (1 - D E P R_{O C}) O C_{i, t - 1} + \frac{S G A_{i, t}}{C P I_{t}},

(1)where DEPR_OC is the depreciation rate, and CPI_t is the consumer price index. We need the initial stock and a depreciation rate to implement the law of motion as described above. We set the initial stock of OC equal to:

O C_{i, 0} = \frac{S G A_{i, 1}}{g + D E P R_{O C}},

(2)where g is the 10-year average (two-digit SIC) industry real growth rate of SG&A. SGA_i,1 is firm i’s first-year SG&A expenses in Compustat with non-missing data. We set subsequent missing values of firm i’s SG&A expenses to 0. Finally, we scale OC by total assets.

Whistleblowing Sample

Following Bowen et al. (2010), we collect our sample of external WB allegations from two sources. The first is the press sample. We manually collect WB events related to financial misconduct reported in the media over 1990–2018 by searching LexisNexis for each combination of the following two groups of keywords: (1) “whistle,” “WB,” “whistle-blowing,” “whistleblower,” and “whistle-blower” and (2) “financial,” “accounting,” “reporting,” “fraud,” and “accounting fraud.” We then carefully read each media report and retain only those related to financial misconduct. We begin with 231 media-reported WB cases.

The second source is the OSHA sample. It consists of WB discrimination cases gathered from written requests to OSHA under the FOIA. A whistleblower can report perceived workplace discrimination to OSHA (e.g., demotions and terminations), or that an employer or anyone in the organization has retaliated against a whistleblower for voicing allegations of financial misconduct, such as a violation of SOX or the Consumer Financial Protection Act (CFPA, Title X of the Dodd–Frank Act). Although OSHA handles other types of WB allegations, we obtain data solely for allegations of financial misconduct from those who filed for protection under SOX or the CFPA over 2002–2018. Our OSHA sample consists of WB cases protected by §806 of SOX or Title X of the Dodd–Frank Act. We begin with 3,546 OSHA cases, and subsequently exclude 1,662 that have no CUSIP data in Compustat and 81 that are duplicates in the same firm-year.

We require that firms have non-missing data in Compustat, the Center for Research in Security Prices (CRSP), and I/B/E/S databases for year t − 1 for PSM and for our main analyses, where t is the year of the WB event. We exclude another 1,092 observations at this step, leaving us with a final sample of 942 WB cases from 518 unique firms for the 1989–2020 period.

Table 1 presents the distribution by year, industry, and type (i.e., press vs. OSHA). The sample covers WB cases across multiple industries, and is distributed relatively evenly across the sample period.

Table 1.

Sample Distribution: Whistleblowing Allegations by Year, Industry, and Type.

Year Distribution
	Press	OSHA
1990	2	NA
1991	3	NA
1992	3	NA
1993	2	NA
1994	3	NA
1995	2	NA
1996	14	NA
1997	5	NA
1998	1	NA
1999	2	NA
2000	3	NA
2001	5	NA
2002	5	21
2003	2	37
2004	5	70
2005	2	79
2006	1	73
2007	2	65
2008	4	63
2009	4	81
2010	2	41
2011	0	49
2012	3	45
2013	8	43
2014	7	44
2015	1	51
2016	3	42
2017	2	28
2018	5	9
All years	101	841

Industry Distribution
	Press	OSHA
Agriculture	0	3
Food products	3	14
Candy and soda	1	7
Beer and liquor	1	12
Recreation	1	1
Entertainment	0	5
Printing and publishing	0	5
Consumer goods	0	7
Apparel	1	4
Healthcare	15	11
Medical equipment	4	37
Pharmaceutical products	13	44
Chemicals	2	10
Textiles	0	1
Construction materials	1	10
Construction	1	25
Steelworks, etc.	0	3
Machinery	1	24
Electrical equipment	0	7
Automobiles and trucks	1	12
Aircraft	4	25
Shipbuilding, railroad equipment	1	0
Defense	1	3
Precious metals	0	1
Metallics and industrial metal mining	0	2
Coal	0	1
Petroleum and natural gas	14	19
Utilities	1	38
Communication	4	55
Personal services	1	15
Business services	5	150
Computers	0	23
Electronic equipment	2	35
Measuring, control equipment	0	15
Business supplies	0	4
Transportation	0	34
Wholesale	2	19
Retail	13	64
Restaurants, hotels, motels	0	13
Banking	0	8
Insurance	2	24
Real estate	0	5
Trading	1	22
Other	5	24
All industries	101	841

This table reports the distribution by year, Fama and French (1997) 48-industry classification, and type of whistleblowing incident (press or OSHA) for the sample of 942 whistleblowing allegations between 1990 and 2018.

Propensity Score-Matched Sample

One concern related to making inferences about changes in OC investment following WB events is that the events may not be exogenous. Rather, they may be related to non-random activities linked to the allegations. Following prior studies (Bowen et al., 2010; Wilde, 2017), we use a propensity score-matched control sample to mitigate these concerns. This matching approach allows us to attribute observed outcomes to WB instead of to other allegation-related firm behaviors.

The PSM design can mitigate any misspecification of the functional form underlying the association between the variable of interest and a given outcome (Armstrong et al., 2010). We include the determinants of WB and OC in the matching model. We adopt a DiD design with the WB firms (WB, treatment sample) and the propensity score-matched sample (PSM control sample) that exhibit appropriate covariates in the pre-WB period.

Following Bowen et al. (2010), our determinants of WB (CONTROLS_WB) include: capital market pressure (CM_PRESSURE); sales growth (SALEGROWTH); unclear communication channels (COM); employee downsizing (DOWNSIZE); membership in an industry with monetary incentives to report misconduct (QUITAM); firm size (SIZE); internal control weaknesses (ICW); and external monitoring (ANALYST).

We also draw on prior research (Eisfeldt & Papanikolaou, 2013, 2014; Li et al., 2018) to identify factors associated with investment in OC (CONTROLS_OC). Accordingly, we control for investment in research and development (R&D), investment in fixed assets (CAPEX), cash holdings (CASH), dividend payments (DIV), share repurchase expenditures (REPO), past performance (ROA), and prior level of investment in OC (LAGOC).

To compute propensity scores, we estimate the following logistic regression:

{P r o b (W B)}_{i, t} = α_{0} + α_{j} {C O N T R O L S}_{W B i, t - 1} + α_{k} {C O N T R O L S}_{O C i, t - 1} + ε_{i, t},

(3)where WB equals 1 for WB firms in the year of the event, and 0 otherwise; CONTROLS_WB is a vector of WB determinants; and CONTROLS_OC is a vector of OC determinants, as discussed previously. We estimate model (3), and require the matched firm to be in the same two-digit SIC industry-year. We measure the independent variables in model (3) in the year before the event (t − 1) in order to match control firms to WB firms based on the determinants of allegations and OC prior to the event. This matching approach improves the power of the tests to detect subsequent changes in OC (Armstrong et al., 2010; Wilde, 2017).

Next, we refer to McFadden’s (1974) pseudo-R² in order to assess the goodness of fit of the logistic model. McFadden (1974) suggests that a pseudo-R² between 0.20 and 0.40 represents “an excellent fit.” Our model (3) estimation yields a McFadden’s pseudo-R² of 23.92%, suggesting satisfactory goodness of fit in predicting the likelihood of WB events. We also validate the predictive power of our model by using the area under the receiver operating characteristic curve (ROC). Rice and Harris (2005) find that an ROC value of 0.70 or higher indicates a satisfactory model fit. The value for our model (3) estimation is 0.89. The PSM-balanced sample consists of 942 external WB observations and 942 PSM control observations.

Table 2 presents descriptive statistics for the propensity score-matched sample, divided into pre- and post-allegation periods. Panel A shows descriptive statistics for the pre-allegation period year (t − 1); Panel B describes the post-allegation period year (t + 1). We exclude year t from the analyses to ensure a clean comparison between periods.

Table 2.

Summary Statistics.

	WB Treatment Sample (N = 942)					PSM Control Sample (N = 942)
Variable	Mean	Std	Q1	Median	Q3	Mean	Std	Q1	Median	Q3
Panel A: Descriptive statistics for the whistleblowing (WB) sample and the PSM control sample in year t − 1
CONTROLS_WB
CM_PRESSURE	60.027	56.463	15.833	37.750	98.833	59.390	59.914	16.666	34.750	86.667
SALEGROWTH	62.729	69.544	13.000	37.000	90.000	61.066	75.118	11.000	30.500	81.000
COM	64.095	58.896	16.333	40.333	102.833	64.193	65.027	16.833	36.667	94.167
DOWNSIZE	61.570	68.236	12.000	36.000	89.000	60.079	74.048	11.000	27.000	80.000
QUITAM	0.162	0.369	0.000	0.000	0.000	0.162	0.240	0.000	0.000	0.000
SIZE	8.888	2.314	7.401	9.034	10.572	6.380	2.109	4.937	6.336	7.820
ICW	0.011	0.012	0.006	0.010	0.016	0.014	0.011	0.006	0.011	0.017
ANALYST	4.454	13.818	0.000	0.000	12.000	4.754	9.159	0.000	0.000	6.000
CONTROLS_OC
R&D	0.028	0.057	0.000	0.000	0.035	0.026	0.063	0.000	0.000	0.020
CAPEX	0.043	0.044	0.016	0.031	0.057	0.055	0.058	0.018	0.036	0.074
CASH	0.137	0.151	0.018	0.087	0.185	0.127	0.163	0.018	0.056	0.172
DIV	0.016	0.029	0.000	0.007	0.024	0.012	0.043	0.000	0.000	0.014
REPO	0.037	0.129	0.000	0.006	0.042	0.013	0.035	0.000	0.000	0.008
ROA	0.042	0.137	0.015	0.049	0.093	0.013	0.152	−0.005	0.031	0.070
LAGOC	0.098	0.218	0.003	0.011	0.051	0.496	0.970	0.015	0.094	0.455
Dependent variables of interest
OC	0.127	0.287	0.004	0.012	0.066	0.571	1.047	0.017	0.114	0.600
Panel B: Descriptive statistics for the whistleblowing (WB) sample and the PSM control sample in year t + 1
CONTROLS_WB
CM_PRESSURE	62.020	57.741	16.00	39.333	94.667	66.545	69.978	16.833	38.000	99.000
SALEGROWTH	66.497	72.078	14.000	39.000	96.000	64.661	74.976	12.000	34.000	90.000
COM	67.355	61.595	17.333	42.167	105.500	67.401	67.826	17.167	39.500	100.667
DOWNSIZE	65.226	70.699	14.000	36.000	99.000	63.470	74.787	12.000	31.000	87.000
QUITAM	0.162	0.368	0.000	0.000	0.000	0.166	0.248	0.000	0.000	0.000
SIZE	8.827	2.322	7.336	8.967	10.477	6.316	2.097	4.902	6.200	7.800
ICW	0.011	0.011	0.006	0.010	0.015	0.013	0.012	0.005	0.011	0.017
ANALYST	4.886	14.787	0.000	1.000	12.000	4.971	9.979	0.000	0.000	6.000
CONTROLS_OC
R&D	0.028	0.055	0.000	0.000	0.039	0.027	0.069	0.000	0.000	0.021
CAPEX	0.045	0.048	0.018	0.033	0.059	0.060	0.068	0.018	0.040	0.076
CASH	0.138	0.155	0.030	0.083	0.188	0.126	0.166	0.016	0.061	0.171
DIV	0.016	0.031	0.000	0.008	0.023	0.017	0.115	0.000	0.000	0.013
REPO	0.032	0.064	0.000	0.004	0.038	0.014	0.043	0.000	0.000	0.007
ROA	0.045	0.112	0.014	0.049	0.093	0.021	0.311	−0.003	0.034	0.076
LAGOC	0.138	0.430	0.004	0.012	0.055	0.525	1.113	0.015	0.094	0.455
Dependent variable of interest
OC	0.201	0.627	0.005	0.015	0.083	0.579	1.207	0.016	0.114	0.579

Panel A (Panel B) presents descriptive statistics for the main variables of the whistleblowing sample and the PSM matched sample in year t − 1 (year t + 1), where year t is the whistleblowing event year. Appendix provides the details of variable construction and definitions.

Changes in Organizational Capital

We adopt a DiD approach to examine changes in OC investment from the pre- to post-allegation periods for the treatment versus PSM control firms. DiD estimation is appropriate when the event period is common across observations (Chen et al., 2015; Wilde, 2017). In our setting, the common event period is the year of the (pseudo-) allegation for the matched and control firms (despite the fact that individual WB events occur in different years). Following Caliendo and Kopeinig (2008) and Roberts and Whited (2013), we examine within-matched-pair variation to remove selection bias from non-random events and time-invariant unobservable attributes. This approach also addresses concerns that changes in OC, common to both types of firms, may drive the observed behaviors.

We estimate the following regression model:

O C_{i, t} = β_{1} W B_{i} + β_{2} P O S T_{i} + β_{3} W B_{i} \times P O S T_{i} + β_{J} C O N T R O L S_{i, t} + F E + ɛ_{i, t},

(4)where OC is the OC measure; WB is an indicator variable coded as 1 if the firm-year is involved in a WB event, and 0 otherwise; POST is an indicator variable coded as 1 for year t + 1, where year t is the WB event year for WB firms and the pseudo-event year for control firms, and 0 otherwise; and CONTROLS is a set of control variables, namely, firm size (SIZE), growth (MTB), leverage (LEV), investment in R&D (R&D), investment in fixed assets (CAPEX), cash holdings (CASH), dividend payments (DIV), share repurchase expenditures (REPO), and performance (ROA). We adjust standard errors for clustering by firm (Petersen, 2009). We include year and Fama and French (1997) industry fixed effects. To aid in interpreting the economic significance, we standardize all continuous regression variables. Consistent with our hypothesis, we expect β₃ to be positive.

Empirical Results

Changes in Organizational Capital: Main Results

Table 3 presents the differences in means and medians of OC between the pre- and post-WB periods for WB firms in column (1), and between WB firms and PSM control firms in column (2). During the pre-allegation period, we observe that WB firms have a significantly lower OC than PSM control firms (−0.444, p-value <.01), with mean OC values of 0.127 and 0.571, respectively. During the post-allegation period, WB firms increase their OC to 0.201, a significant increase from the pre-allegation level (the difference is 0.074; p-value <.01). This exceeds the change in OC for PSM firms (the difference is 0.066; p-value <.05). The median OC differences show similar results. Thus far, the results from our univariate analysis align with our expectation that WB firms increase investment in OC during the post-allegation period.

Table 3.

Changes in Organizational Capital in Pre- and Post-whistleblowing Allegation Periods.

	Whistleblowing (WB) Firms[1]		PSM Control Firms[2]		Difference WB − Control[1]–[2]
	Mean	Median	Mean	Median	Mean	Median
Pre-period	0.127	0.012	0.571	0.114	−0.444***	−0.102***
Pre-period	0.127	0.012	0.571	0.114	(t-stat. = −12.53)	(z-stat. = −15.24)
Post-period	0.201	0.015	0.579	0.114	−0.378***	−0.100***
Post-period	0.201	0.015	0.579	0.114	(t-stat. = −7.96)	(z-stat. = −12.28)
Difference	0.074***	0.003***	0.008	0.000	0.066**	0.002***
Difference	(t-stat. = 3.24)	(z-stat. = 2.74)	(t-stat. = 0.16)	(z-stat. = 0.34)	(t-stat. = 1.90)	(z-stat. = 9.02)

This table presents the differences in means (matched-pair t-tests) and medians (Wilcoxon signed-rank tests) of organizational capital (OC) for the pre- and post-whistleblowing periods of whistleblowing (WB) and control firms.

***, **, and * indicate statistical significance at the 0.01, 0.05, and 0.10 levels (one-tailed), respectively.

Appendix provides detailed variable descriptions.

Table 4 reports results from the changes in OC regression analyses. The coefficient on WB × POST in columns (1) and (2) is positive and significant (p-value <.05). The result provides evidence that WB firms increase investment in OC during the post-allegation period. Specifically, the coefficient on WB × POST is 0.022 in column (1), which indicates an average increase of 2.20% of a standard deviation in OC more than their PSM control firm in the post-allegation period. Using WB firms in year t − 1 as the benchmark, we estimate that the subsequent increase is approximately 5% of the mean (0.022×0.287/0.127), which translates to about $45.74 million (0.022×0.287×e^8.888). In addition, the control variables show coefficient estimates that are consistent with prior research (Eisfeldt & Papanikolaou, 2013; Gao et al., 2021).

Table 4.

Changes in Organizational Capital Around Whistleblowing Allegations.

		Sample Period: Year t − 1 and Year t + 1
		Dependent Variable OC
		Coeff. (t-stat)	Coeff. (t-stat)
	Expected Sign	[1]	[2]
WB		−0.032	−0.165***
WB		(−1.41)	(−6.88)
POST		−0.015	−0.017
POST		(−1.26)	(−1.42)
WB × POST	+	0.022**	0.026***
WB × POST		(1.90)	(2.26)
SIZE		−0.210***	−0.180***
SIZE		(−6.06)	(−5.24)
MTB		−0.009	−0.011
MTB		(−0.77)	(−0.69)
LEV		−0.052*	−0.060**
LEV		(−1.86)	(−2.13)
EMPGROWTH		0.110***	0.113***
EMPGROWTH		(2.78)	(2.81)
R&D		0.118***	0.124***
R&D		(2.69)	(2.78)
CAPEX		−0.037	−0.004
CAPEX		(−1.55)	(−0.18)
CASH		0.091***	0.068*
CASH		(2.54)	(1.84)
DIV		0.067*	0.075*
DIV		(1.63)	(1.65)
REPO		0.014	−0.002
REPO		(0.68)	(−0.09)
ROA		−0.076*	−0.070
ROA		(−1.72)	(−1.61)
Industry fixed effects		Yes	No
Year fixed effects		Yes	No
N		3,768	3,768
Adjusted R²		27.62%	17.88%

This table presents results from difference-in-difference regressions of organizational capital (OC) around whistleblowing allegations in years t − 1 and t + 1 for the PSM sample of 3,768 observations (942 PSM pairs over two years). All continuous variables are standardized to a mean of 0 and a standard deviation of 1. p-values are based on standard errors clustered by firm.

***, **, and * Indicate statistical significance at the 0.01, 0.05, and 0.10 levels, respectively.

p-values are one-sided for variables with directional predictions. Appendix provides detailed variable descriptions.

Alternative Measures of Organizational Capital: Robustness Check

We note that concerns may arise when using SG&A to measure OC, because accounting treatments of SG&A expenses can vary across industries. They mainly reflect OC-related investments, such as employees’ training, welfare, and benefits, and communication and technology expenses. But they may also include unrelated expenses, like advertising, R&D, and legal fees. This may introduce noise into the OC measure. We address this issue by conducting a battery of robustness checks using multiple alternative OC measures.

First, following Li et al. (2018), we use the industry median-adjusted ratio of OC to total assets (IND_ADJ_OC). Second, we use the ratio of SG&A expenses to total assets (SGA_OC). Third, to account for the possibility that higher OC investment may stem from increased legal expenses related to WB allegations, we subtract legal expenses from SG&A (LITIEXP_ADJ_OC). Finally, for our fourth measure, we exclude litigation, advertising, and R&D expenses from SG&A expenses before calculating OC (OTHEREXP_ADJ_OC).

The regression results using these alternative measures are in columns (1)–(4) of Table 5. We find that the coefficient on WB × POST is positive and significant at the 5% level in all four models. This indicates that WB firms increase their OC investment during the post-allegation period. Thus, our main findings are robust.

Table 5.

Changes in Organizational Capital Around Whistleblowing Allegations: Alternative Organizational Capital Measures.

		Sample Period: Year t − 1 and Year t + 1
		Dependent Variable IND_ADJ_OC	Dependent Variable SGA_OC	Dependent Variable LITIEXP_ADJ_OC	Dependent Variable OTHEREXP_ADJ_OC
	Expected Sign	Coeff. (t-stat)	Coeff. (t-stat)	Coeff. (t-stat)	Coeff. (t-stat)
	Expected Sign	[1]	[2]	[3]	[4]
WB		−0.155***	−0.070***	−0.032	−0.032
WB		(−6.27)	(−2.45)	(−1.39)	(−1.40)
POST		−0.017	−0.014	−0.013	−0.013
POST		(−1.40)	(−1.51)	(−1.11)	(−1.06)
WB × POST	+	0.024**	0.023**	0.022**	0.022**
WB × POST		(2.16)	(1.90)	(2.00)	(1.95)
Controls		Yes	Yes	Yes	Yes
Industry fixed effects		Yes	Yes	Yes	Yes
Year fixed effects		Yes	Yes	Yes	Yes
N		3,768	3,768	3,768	3,768
Adjusted R²		11.25%	14.44%	16.96%	15.55%

This table presents results from difference-in-difference regressions of alternative organizational capital (OC) measures around whistleblowing allegations in years t ‒ 1 and t + 1 for the PSM sample of 3,768 observations (942 PSM pairs over two years). IND_ADJ_OC is organizational capital of firm i minus the two-digit SIC industry median organizational capital. SGA_OC is organizational capital measured as the ratio of SG&A expenses to total assets. LITIEXP_ADJ_OC is organizational capital measured as capitalized SG&A expenses minus litigation expenses (SETA) scaled by total assets, using the perpetual inventory method as specified in Li et al. (2018). OTHEREXP_ADJ_OC is organizational capital measured as capitalized SG&A expenses minus litigation expenses (SETA), advertising expenses (XAD), and research and development expenses (XRD) scaled by total assets, using the perpetual inventory method as specified in Li et al. (2018). All variables are standardized to a mean of 0 and a standard deviation of 1. p-values are based on standard errors clustered by firm.

***, **, and * Indicate statistical significance at the 0.01, 0.05, and 0.10 levels, respectively.

p-values are one-sided for variables with directional predictions. Appendix provides detailed variable descriptions.

Qualitative Measures of Organizational Capital: Robustness Check

Another concern is that OC does not directly measure the costs associated with improving a firm’s internal communications environment. It can be challenging to identify the exact amounts invested in the various dimensions of OC, such as WB hotlines, corporate citizenship programs, or employee training initiatives. Such information is not typically publicly disclosed. To address this, we use qualitative disclosures of OC dimensions related to reinforcing employees’ knowledge of internal WB reporting systems, improving corporate citizenship, business ethics, and companywide anti-bribery, corruption, and fraud policies and activities.⁴

We obtain four different qualitative OC dimensions: ETHICS, INVOLVEMENT, TRAINING, and ANTI-FRAUD from Refinitiv’s environmental, social, and governance (ESG) database. This database covers over 12,500 public and private companies globally, including more than 3,800 U.S. companies. The data are available from 2003 onward, with dimension scores ranked by industry. These dimensions fall under the workforce, human rights, and community categories within the social pillar of Refinitiv’s ESG scores.⁵

According to Refinitiv, the business ethics (ETHICS) dimension measures a company’s internal communication tools and policies for improving business ethics. Employee involvement in the internal reporting system (INVOLVEMENT) measures the internal communication tools and policies related to whistleblowers, ombudsmen, suggestion boxes, hotlines, newsletters, websites, etc., that are designed to improve corporate citizenship. Employee training (TRAINING) measures the training on internal policies related to whistleblowers, ombudsmen, suggestion boxes, hotlines, newsletters, websites, etc. And the anti-bribery, corruption, and fraud (ANTI-FRAUD) dimension focuses on internal communication tools and policies to prevent bribery and corruption.

We aggregate these four dimensions into a composite measure of qualitative OC (OC_QUALITATIVE), and then merge it with our main sample. This results in 556 observations from 139 WB firms and their matched observations. We estimate the association between the original OC measure and the qualitative OC measure using the following regression model:

O C_{i, t} = β_{0} + β_{1} O C_Q U A L I T A T I V E_{i} + F E + ɛ_{i, t} .

(5)

The (unreported) regression results show that the coefficient on OC_QUALITATIVE is positive and significant (the coefficient is 0.309; p-value <.01). This suggests that our original measure of OC is strongly correlated with the underlying constructs of OC that are related to communication with employees to improve business ethics, increasing employee involvement, enhancing training, and strengthening communication about anti-fraud activities. We then obtain the expected value of the dependent OC (OC_HAT) from model (5) above.

Table 6 reports the results from re-estimating model (4) using alternative dependent variables: OC_HAT and OC_QUALITATIVE, along with its subcomponents. The findings reinforce the main findings that WB firms increase investments in OC, particularly in areas such as business ethics, employee involvement, training, and anti-fraud efforts.

Table 6.

Changes in Organizational Capital Around Whistleblowing Allegations: Qualitative Measures.

		Sample Period: Year t − 1 and Year t + 1
		Dependent Variable	Dependent Variable	Dependent Variable	Dependent Variable	Dependent Variable	Dependent Variable
		OC_HAT	OC_QUALITATIVE	ETHICS	INVOLVEMENT	TRAINING	ANTI-FRAUD
	Expected Sign	Coeff. (t-stat)	Coeff. (t-stat)	Coeff. (t-stat)	Coeff. (t-stat)	Coeff. (t-stat)	Coeff. (t-stat)
	Expected Sign	[1]	[2]	[3]	[4]	[5]	[6]
WB		0.176**	0.157*	0.121	0.157*	0.160**	0.126
WB		(2.13)	(1.83)	(1.44)	(1.76)	(2.15)	(1.50)
POST		−0.023	−0.042	−0.037	−0.042	−0.034	−0.037
POST		(−0.93)	(−1.58)	(−1.31)	(−1.52)	(−1.29)	(−1.32)
WB × POST	+	0.100***	0.104***	0.102***	0.122***	0.061*	0.084**
WB × POST		(2.26)	(2.23)	(2.22)	(2.43)	(1.50)	(1.82)
Controls		Yes	Yes	Yes	Yes	Yes	Yes
Industry fixed effects		Yes	Yes	Yes	Yes	Yes	Yes
Year fixed effects		Yes	Yes	Yes	Yes	Yes	Yes
N		556	556	556	556	556	556
Adjusted R²		30.70%	27.92%	24.83%	25.03%	24.92%	24.30%

This table presents results from difference-in-difference regressions of qualitative organizational capital (OC) measures around whistleblowing allegations in years t − 1 and t + 1. OC is constructed based on qualitative measures of a firm’s communication tools related to business ethics improvement (ETHICS), good corporate citizenship (INVOLVEMENT), employee training and development (TRAINING), and bribery and corruption avoidance (ANTI-FRAUD). All variables are standardized to a mean of 0 and a standard deviation of 1. p-values are based on standard errors clustered by firm.

***, **, and * Indicate statistical significance at the 0.01, 0.05, and 0.10 levels, respectively.

p-values are one-sided for variables with directional predictions. Appendix provides detailed variable descriptions.

Entropy-Balanced Sample: Robustness Check

The PSM approach is effective at producing appropriate covariate matching. However, in order to mitigate concerns about PSM’s sensitivity to design choice, we also use entropy balancing. This approach mitigates differences in observable covariates across the treatment and control groups (Hainmueller, 2012). We balance the sample using the same set of WB allegation determinants used in the PSM approach. We then equalize the first (mean), second (variance), and third (skewness) moments of each covariate distribution using a tolerance level of 0.015 for convergence. Consistent with our previous findings, WB firms exhibit higher levels of OC investment post-allegation.⁶

Additional Analyses

Determinants of Organizational Capital Improvement Decisions

The results from the previous sections indicate that WB firms on average increase their investment in OC during the post-allegation period. However, in our sample, only 75.16% of firms increased their OC from year (t − 1) to (t + 1). To explore which factors influence this decision, we examine several key determinants, including labor and shareholder influences, pre-allegation OC levels, WB case credibility, and CEO incentives.

To this end, we estimate the following regression model:

P r o b (I N C R E A S E_O C_{i, t}) = β_{1} L A B O R_{i, t} + β_{2} N E G C A R_{i} + β_{3} L A G O C_{i, t} + β_{4} M E R I T_{i} + β_{5} L T I O_{i, t} + β_{6} S T I O_{i, t} + β_{7} C E O_O W N_{i, t} + β_{8} V E G A_{i, t} + β_{9} C E O_C H A I R_{i, t} + β_{10} F E M A L E_C E O_{i, t} + F E + ɛ_{i, t},

(6)where the dependent variable is a dummy variable that equals 1 if the WB firm increases OC from year (t − 1) to (t + 1), and 0 otherwise.

Labor-intensive firms by nature require larger investments in human capital-related components of OC (such as communication systems, workforce training, and worker welfare). Thus, we expect labor-intensive firms (measured using LABOR) to be more likely to increase OC investment in response to WB allegations. Furthermore, negative market reactions to WB allegations can cause significant losses for shareholders, prompting firms to act. We expect WB firms that experience negative market reactions (NEGCAR) to be more likely to increase OC.

Insufficient prior investment in OC can lead to lax internal controls, ineffective communications, and low employee morale, which may increase the odds of financial misconduct and subsequent external WB. Therefore, we expect WB firms with lower OC during the pre-allegation period (LAGOC) to be more likely to increase OC investment. Research has also found that some WB allegations reflect employees’ reckless assessments of a situation (Berger & Lee, 2022). Thus, credible allegations (MERIT) likely indicate OC deficiencies, and are more likely to motivate managers to increase OC investment. We expect that firms facing credible WB allegations will be more likely to increase investment in OC post-allegation. The sample size decreases because credibility information is only available for OSHA WB allegations.

Long-term institutional investors enhance managerial decision-making through their monitoring role. They help restrain financial fraud and improve investment decisions (Gaspar et al., 2005; Harford et al., 2018). Therefore, WB firms with a higher percentage of long-term institutional ownership are more likely to increase investment in OC. On the other hand, short-term institutional investors tend to prioritize immediate returns over long-term investments (Attig et al., 2013; Gaspar et al., 2005), thereby undermining their effectiveness in monitoring managers.

Following Gaspar et al. (2005), we use investor-level portfolio information from the 13F holdings compiled by Refinitiv to compute long-term institutional ownership (LTIO) and short-term institutional ownership (STIO). We expect to find a positive (negative) and significant coefficient on LTIO (STIO).

We also examine the impact of CEO factors such as equity ownership (CEO_OWN), and risk-taking incentives such as compensation’s sensitivity to stock return volatility (VEGA). This is because CEOs with significant ownership stakes should be more motivated to address the deficiencies highlighted by WB in order to protect their personal financial interests. Shen and Zhang (2013) find that firms with increased VEGA are more likely to increase R&D investments. Higher VEGA could also influence increases in OC investment, because higher sensitivities motivate CEOs to undertake riskier investments (Coles et al., 2006; Core & Guay, 2002). Thus, we expect to find a positive and significant coefficient on CEO_OWN and VEGA.

Prior literature suggests that CEO duality (where the CEO also serves as the chair of the board) can reduce a board’s CEO monitoring ability. This increases the likelihood of financial misreporting (Dechow et al., 1996; Doyle et al., 2007), which can result in suboptimal investment decisions (Masulis et al., 2007; Yermack, 1996). We therefore posit that CEO duality (CEO_CHAIR) will be associated with a lower likelihood that a CEO will address OC deficiencies.

Finally, female CEOs are often associated with ethical characteristics such as empathy and integrity (Chun, 2005). They tend to be more trustworthy and more compliant with rules and regulations (Baldry, 1987; Fallan, 1999). Therefore, we examine the impact of CEO gender. We posit that female CEOs (FEMALE_CEO) are more likely to increase OC post-allegation. Table 7 reports the logistic regression results from estimating model (6) with and without fixed effects.

Table 7.

Determinants of Increased Organizational Capital.

		Sample Period: Year t − 1 and Year t + 1
		Dependent Variable INCREASE_OC
	Expected Sign	Coeff.	Coeff.	Coeff.	Coeff.	Coeff.	Coeff.	Coeff.	Coeff.
	Expected Sign	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
LABOR	+	0.393***	0.380***	0.242***	0.250***	0.353***	0.350***	0.244***	0.249***
LABOR		(81.08)	(66.60)	(18.21)	(18.72)	(59.88)	(53.48)	(17.46)	(17.77)
NEGCAR	+	0.309***	0.259***	0.098**	0.098**	0.259***	0.231***	0.091*	0.092*
NEGCAR		(49.94)	(30.17)	(2.88)	(2.82)	(31.85)	(22.45)	(2.35)	(2.36)
LAGOC	‒	−0.093**	−0.115***	−0.099**	−0.103**	−0.118***	−0.122***	−0.117**	−0.123**
LAGOC		(4.10)	(4.73)	(3.27)	(3.40)	(6.00)	(4.99)	(4.11)	(4.31)
MERIT	+		0.145***	0.042	0.042		0.126***	0.050	0.049
MERIT			(7.40)	(0.50)	(0.49)		(5.45)	(0.66)	(0.62)
LTIO	+			0.140***	0.163***			0.121***	0.147***
LTIO				(11.23)	(7.67)			(7.84)	(6.12)
STIO	‒			0.063	0.070			0.055	0.058
STIO				(1.39)	(1.54)			(0.95)	(0.95)
CEO_OWN	+				−0.035				−0.020
CEO_OWN					(0.20)				(0.07)
VEGA	+				0.034				−0.016
VEGA					(0.29)				(0.06)
CEO_CHAIR	‒				−0.025				−0.023
CEO_CHAIR					(0.12)				(0.09)
FEMALE_CEO	+				0.016				−0.002
FEMALE_CEO					(0.05)				(0.00)
Industry fixed effects		No	No	No	No	Yes	Yes	Yes	Yes
Year fixed effects		No	No	No	No	Yes	Yes	Yes	Yes
N		812	725	725	725	812	725	725	725
McFadden’s pseudo-R²		30.86%	41.28%	44.53%	44.64%	35.73%	46.06%	47.99%	48.07%

This table reports results from the logistic regressions of the decision to increase organizational capital in year t + 1 of WB firms. INCREASE_OC equals 1 if OC in year t + 1 is greater than that in year t − 1, and 0 otherwise. All variables are standardized to a mean of 0 and a standard deviation of 1. Wald chi-squares are in parentheses.

***, **, and * Indicate statistical significance at the 0.01, 0.05, and 0.10 levels (one-tailed), respectively.

Appendix provides detailed variable descriptions.

LABOR has the largest standardized coefficient. WB firms with higher labor intensity are more likely to increase investment in OC post-allegation. This suggests that employees play an important role in a WB firm’s decision to invest in OC, which reinforces the prediction of prosocial organizational behavior theory. We also find that WB firms with negative stock market reactions, higher long-term institutional ownership, and lower pre-allegation OC levels are more likely to increase OC investment even after controlling for industry and year fixed effects.

Note further that the positive coefficient on MERIT in columns (2) and (6) of Table 7 suggests that credible WB cases lead to increased OC investment. However, when we include both LTIO and MERIT, MERIT remains positive but becomes statistically insignificant. Therefore, credibility is not the primary driver of increased OC investment. This finding complements Kuang et al. (2021), who find that auditors charge WB firms higher fees regardless of WB case credibility. On the other hand, factors such as short-term institutional investors, CEO ownership, CEO risk-taking incentives, CEO duality, and CEO gender do not significantly influence the decision to increase OC post-allegation.

In sum, a firm’s decision to increase OC investment reflects a strategic response aimed at enhancing internal processes. This approach is likely driven by the interests of key stakeholders like long-term institutional investors and employees. It underscores the importance of adopting long-term strategic planning and aligning employee benefits with the company’s overarching objectives to cultivate a responsive and adaptive organizational culture.

Impact on Future External Whistleblowing Incidence

We next examine the relation between increasing OC and future WB incidences post-allegation.

The results in Panel A of Table 8 show that WB firms that increase OC in year t + 1 (INCREASE_OC = 1) have a significantly lower likelihood of future WB reports in year t + 2 than those that do not (INCREASE_OC = 0) (20.93% compared to 25.23%; p-value of the difference = 0.09). Thus, the decision by WB firms to increase OC post-allegation is associated with a lower probability of future WB.

Table 8.

Changes in Organizational Capital and Future Whistleblowing Incidence.

Panel A: Actual Whistleblowing Incidence
	WB Allegations in Year t + 2
INCREASE_OC_t+1 = 1	20.93%
INCREASE_OC_t+1 = 0	25.53%
Difference	4.61%*
p-value	(0.09)

Panel B: Correlation Coefficients Between INCREASE_OC and WBPROB
	WBPROB _t+2
INCREASE_OC _t+1	−0.232***
p-value	(<0.01)
	N = 153

Panel C: t Test Comparing WB Probability
	WBPROB _t+2
INCREASE_OC_t+1 = 1	11.68%
	N = 38
INCREASE_OC_t+1 = 0	19.47%
	N = 115
Difference	7.80%***
p-value	(<0.01)

This table reports test results for the impact of the decision to increase OC in year t + 1 (INCREASE_OC_t+1) on WB incidence in year t + 2. Panel A presents the likelihood of actual WB incidents. Panel B presents the correlation coefficient between the indicator of increase in OC in year t + 1 (INCREASE_OC_t+1) and the expected probability of a WB allegation in year t + 2 (WBPROB_t+2) estimated from model (3). Panel C presents the t test results comparing the expected whistleblowing probability in year t + 2 (WBPROB_t+2) across the WB firms that increased OC investment in year t + 1 (INCREASE_OC_t+1 = 1) and those that did not (INCREASE_OC_t+1 = 0). INCREASE_OC equals 1 if OC in year t + 1 is greater than that in year t − 1, and 0 otherwise.

***, **, and * Indicate statistical significance at the 0.01, 0.05, and 0.10 levels (one-tailed), respectively.

Appendix provides detailed variable descriptions.

In Panel B, the correlation coefficient between INCREASE_OC in year t + 1 and the expected probability of whistleblowing (WBPROB) in year t + 2 estimated from model (3) is −0.232 (p-value <.01).

Panel C shows that WB firms that increase OC (INCREASE_OC = 1) post-allegation (year t + 1) face an average expected probability of WB of 11.68%, compared to 19.47% for those that do not (INCREASE_OC = 1), with the difference being statistically significant at the 1% level. These findings underscore the role of OC investment in reducing the likelihood of future WB allegations, providing support for prosocial organizational behavior theory.

Press and OSHA Subsamples

OSHA whistleblower cases offer protections against wrongful termination, discrimination, and retaliation, often reflecting deeper issues than just financial misconduct. Employees may avoid internal reporting due to distrust, lack of feedback, or fear of retaliation. This highlights potential problems with employee trust and the ethical environment.

To examine differences based on the reporting channel, we reestimate model (4) using the press (101 firm-year observations) and OSHA (841 firm-year observations) subsamples. The (unreported) results show that firms in both subsamples increase investment in OC after an allegation. This supports the conclusion that firms respond by enhancing OC investment regardless of the reporting channel.

Brand Capital Investment

We also examine changes in brand capital investment, which refers to consumer awareness of a company’s products or services. Prior research indicates that brand capital significantly impacts a company’s reputation (Belo et al., 2014). Because WB allegations can harm a firm’s reputation, firms may seek to restore consumer trust post-allegation. We measure brand capital investment as accumulated advertising expenses using the perpetual inventory method, similarly to our OC estimation.

Re-estimating model (4) using brand capital investment as the dependent variable, we find that WB firms tend to have higher brand capital investment than control firms. However, we observe no significant change post-allegation. This may suggest that investment in brand capital is seen as “cosmetic,” and does not directly address the internal deficiencies highlighted by WB issues.

Conclusion

Regulators have long viewed WB as an essential mechanism for detecting and exposing corporate misconduct and fraud. As a result, the topic has garnered increasing research attention in recent years. However, several important WB-related issues remain underexplored. Our study addresses one of these gaps: the impact of external WB allegations on OC investment.

External WB typically occurs when internal systems fail to address employees’ concerns, exposing firms to public scrutiny, reputational damage, and litigation risk. These allegations often signal deficiencies in internal controls, communication, and employee engagement—collectively referred to as OC deficiencies. We hypothesize that firms respond to WB allegations by investing in OC in order to rebuild trust and mitigate future misconduct.

Using a difference-in-differences research design with two unique WB datasets, we find that targeted firms significantly increase OC investments post-allegation. These results are robust to propensity score matching, entropy balancing, and alternative OC measures. We also identify key determinants of OC investment, including labor intensity, negative market reactions, pre-existing OC deficiencies, and long-term institutional ownership. These findings underscore the role of employees as internal monitors, and institutional investors as external monitors, in driving improvements.

Our study has several managerial and policy implications. Managers should prioritize addressing OC deficiencies, empowering employees, and fostering an ethical workplace culture to preempt misconduct and reduce the likelihood of external WB. For policymakers, our results affirm the efficacy of WB regulations, such as SOX §806, in exposing misconduct and prompting firms to implement meaningful changes. Enhancing such regulations could further improve their effectiveness.

While we acknowledge the lack of detailed expenditure data to assess the exact cost of OC improvements, our primary focus is on firms’ behavioral responses. Our findings highlight how WB allegations can lead to substantive organizational changes, reflecting the critical role of WB in improving corporate governance and ethical practices.

Footnotes

Author’s Note

We are grateful for helpful comments from two anonymous reviewers, Linda Myers (Editor), Najah Attig, Narjess Boubakri, Andy Call, Jung Ho Choi, Wolfgang Drobetz, Sattar Mansi, Xinming Li, He Wang, Ying Zheng, and participants at the 11^th International Conference of the Journal of International Accounting Research, the 2024 Southwestern Finance Association Annual Meeting, the 2023 American Accounting Association Annual Meeting, the 2023 Financial Management Association Annual Meeting, and the 2022 Southern Finance Association Annual Meeting.

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

Sadok El Ghoul

Omrane Guedhami

Notes

Appendix

References

Andrade

J. A.

(2015). Reconceptualising whistleblowing in a complex world. Journal of Business Ethics, 128(2), 321–335. https://doi.org/10.1007/s10551-014-2105-z

Armstrong

C. S.

Jagolinzer

A. D.

Larcker

D. F.

(2010). Chief executive officer equity incentives and accounting irregularities. Journal of Accounting Research, 48(2), 225–271. https://doi.org/10.1111/j.1475-679x.2009.00361.x

Association of Certified Fraud Examiners [ACFE] . (2020). Report to the nations: 2020 global study on occupational fraud and abuse. https://acfepublic.s3-us-west-2.amazonaws.com/2020-Report-to-the-Nations.pdf

Atkeson

Kehoe

P. J.

(2005). Modeling and measuring organization capital. Journal of Political Economy, 113(5), 1026–1053. https://doi.org/10.1086/431289

Attig

Cleary

El Ghoul

Guedhami

(2012). Institutional investment horizon and investment–cash flow sensitivity. Journal of Banking & Finance, 36(4), 1164–1180. https://doi.org/10.1016/j.jbankfin.2011.11.015

Attig

Cleary

El Ghoul

Guedhami

(2013). Institutional investment horizons and the cost of equity capital. Financial Management, 42(2), 441–477. https://doi.org/10.1111/j.1755-053x.2012.01221.x

Baldry

J. C.

(1987). Income tax evasion and the tax schedule: Some experimental results. Public Finance, 42(3), 357–383.

Belo

Lin

Vitorino

M. A.

(2014). Brand capital and firm value. Review of Economic Dynamics, 17(1), 150–169. https://doi.org/10.1016/j.red.2013.05.001

Berger

P. G.

Lee

(2022). Did the Dodd–Frank whistleblower provision deter accounting fraud? Journal of Accounting Research, 60(4), 1337–1378. https://doi.org/10.1111/1475-679x.12421

10.

Bjørkelo

Einarsen

Nielsen

M. B.

Matthiesen

S. B.

(2011). Silence is golden? Characteristics and experiences of self-reported whistleblowers. European Journal of Work & Organizational Psychology, 20(2), 206–238. https://doi.org/10.1080/13594320903338884

11.

Black

S. E.

Lynch

L. M.

(2005). Measuring organizational capital in the new economy. In Corrado

Haltiwanger

Sichel

(Eds.), Measuring capital in the new economy (pp. 205–236). University of Chicago Press.

12.

Bowen

R. M.

Call

A. C.

Rajgopal

(2010). Whistle-blowing: Target firm characteristics and economic consequences. The Accounting Review, 85(4), 1239–1271. https://doi.org/10.2308/accr.2010.85.4.1239

13.

Brief

A. P.

Motowidlo

S. J.

(1986). Prosocial organizational behaviors. Academy of Management Review, 11(4), 710–725. https://doi.org/10.5465/amr.1986.4283909

14.

Bushman

Chen

Engel

Smith

(2004). Financial accounting information, organizational complexity and corporate governance systems. Journal of Accounting and Economics, 37(2), 167–201. https://doi.org/10.1016/j.jacceco.2003.09.005

15.

Caliendo

Kopeinig

(2008). Some practical guidance for the implementation of propensity score matching. Journal of Economic Surveys, 22(1), 31–72. https://doi.org/10.1111/j.1467-6419.2007.00527.x

16.

Call

A. C.

Martin

G. S.

Sharp

N. Y.

Wilde

J. H.

(2018). Whistleblowers and outcomes of financial misrepresentation enforcement actions. Journal of Accounting Research, 56(1), 123–171. https://doi.org/10.1111/1475-679x.12177

17.

Chen

Tsang

(2015). The effect of mandatory IFRS adoption on international cross-listings. The Accounting Review, 90(4), 1395–1435. https://doi.org/10.2308/accr-50982

18.

Cheng

Bai

Yang

(2019). Ethical leadership and internal whistleblowing: A mediated moderation model. Journal of Business Ethics, 155(1), 115–130. https://doi.org/10.1007/s10551-017-3517-3

19.

Chun

(2005). Ethical character and virtue of organizations: An empirical assessment and strategic implications. Journal of Business Ethics, 57(3), 269–284. https://doi.org/10.1007/s10551-004-6591-2

20.

Coles

J. L.

Daniel

N. D.

Naveen

(2006). Managerial incentives and risk-taking. Journal of Financial Economics, 79(2), 431–468. https://doi.org/10.1016/j.jfineco.2004.09.004

21.

Core

Guay

(2002). Estimating the value of employee stock option portfolios and their sensitivities to price and volatility. Journal of Accounting Research, 40(3), 613–630. https://doi.org/10.1111/1475-679x.00064

22.

Culiberg

Mihelič

K. K.

(2017). The evolution of whistleblowing studies: A critical review and research agenda. Journal of Business Ethics, 146(4), 787–803. https://doi.org/10.1007/s10551-016-3237-0

23.

Daniel

N. D.

Naveen

(2020). Symmetry in pay for luck. Review of Financial Studies, 33(7), 3174–3204. https://doi.org/10.1093/rfs/hhz057

24.

Danielova

Francis

B. B.

Teng

(2023). The effect of organization capital on the cost of bank loans. Journal of Financial and Quantitative Analysis, 58(6), 2579–2616. https://doi.org/10.1017/s0022109022001107

25.

Dechow

P. M.

Sloan

R. G.

Sweeney

A. P.

(1995). Detecting earnings management. The Accounting Review, 70(2), 193–225.

26.

Dechow

P. M.

Sloan

R. G.

Sweeney

A. P.

(1996). Causes and consequences of earnings manipulation: An analysis of firms subject to enforcement actions by the SEC. Contemporary Accounting Research, 13(1), 1–36. https://doi.org/10.1111/j.1911-3846.1996.tb00489.x

27.

Dodd-Frank Wall Street

Reform and Consumer Protection Act [Dodd-Frank], Public law 111-203 [H.R. 4173], (2010). https://www.congress.gov/bill/111th-congress/house-bill/4173/text

28.

Doyle

McVay

(2007). Determinants of weaknesses in internal control over financial reporting. Journal of Accounting and Economics, 44(1–2), 193–223. https://doi.org/10.1016/j.jacceco.2006.10.003

29.

Dyck

Morse

Zingales

(2010). Who blows the whistle on corporate fraud? The Journal of Finance, 65(6), 2213–2253. https://doi.org/10.1111/j.1540-6261.2010.01614.x

30.

Eisfeldt

A. L.

Papanikolaou

(2013). Organization capital and the cross-section of expected returns. The Journal of Finance, 68(4), 1365–1406. https://doi.org/10.1111/jofi.12034

31.

Eisfeldt

A. L.

Papanikolaou

(2014). The value and ownership of intangible capital. The American Economic Review, 104(5), 189–194. https://doi.org/10.1257/aer.104.5.189

32.

Fallan

(1999). Gender, exposure to tax knowledge, and attitudes towards taxation: An experimental approach. Journal of Business Ethics, 18(2), 173–184. https://doi.org/10.1023/a:1005711905297

33.

False Claims Act, 31 U.S.C. §§3729-3733 (1863).

34.

Fama

E. F.

French

K. R.

(1997). Industry costs of equity. Journal of Financial Economics, 43(2), 153–193. https://doi.org/10.1016/s0304-405x(96)00896-3

35.

Gao

Brink

A. G.

(2017). Whistleblowing studies in accounting research: A review of experimental studies on the determinants of whistleblowing. Journal of Accounting Literature, 38(1), 1–13. https://doi.org/10.1016/j.acclit.2017.05.001

36.

Gao

Leung

Qiu

(2021). Organization capital and executive performance incentives. Journal of Banking & Finance, 123, 106017. https://doi.org/10.1016/j.jbankfin.2020.106017

37.

Gaspar

J. M.

Massa

Matos

(2005). Shareholder investment horizons and the market for corporate control. Journal of Financial Economics, 76(1), 135–165. https://doi.org/10.1016/j.jfineco.2004.10.002

38.

Gobert

Punch

(2000). Whistleblowers, the public interest, and the public interest disclosure Act 1998. The Modern Law Review, 63(1), 25–54. https://doi.org/10.1111/1468-2230.00249

39.

Hainmueller

(2012). Entropy balancing for causal effects: A multivariate reweighting method to produce balanced samples in observational studies. Political Analysis, 20(1), 25–46. https://doi.org/10.1093/pan/mpr025

40.

Harford

Kecskés

Mansi

(2018). Do long-term investors improve corporate decision making? Journal of Corporate Finance, 50, 424–452. https://doi.org/10.1016/j.jcorpfin.2017.09.022

41.

Hartman

Desjardins

(2008). Business ethics: Decision-making for personal integrity & social responsibility. McGraw-Hill.

42.

Hoffman

W. M.

Schwartz

M. S.

(2015). The morality of whistleblowing: A commentary on richard T. De George. Journal of Business Ethics, 127(4), 771–781. https://doi.org/10.1007/s10551-014-2186-8

43.

Huang

Teoh

S. H.

Zhang

(2014). Tone management. The Accounting Review, 89(3), 1083–1113. https://doi.org/10.2308/accr-50684

44.

Jones

J. J.

(1991). Earnings management during import relief investigations. Journal of Accounting Research, 29(2), 193–228. https://doi.org/10.2307/2491047

45.

Kuang

Y. F.

Lee

Qin

(2021). Whistleblowing allegations, audit fees, and internal control deficiencies. Contemporary Accounting Research, 38(1), 32–62. https://doi.org/10.1111/1911-3846.12619

46.

Lee

Fargher

(2013). Companies’ use of whistle-blowing to detect fraud: An examination of corporate whistle-blowing policies. Journal of Business Ethics, 114(2), 283–295. https://doi.org/10.1007/s10551-012-1348-9

47.

Lee

Xiao

(2018). Whistleblowing on accountingy-related misconduct: A synthesis of the literature. Journal of Accounting Literature, 41(1), 22–46. https://doi.org/10.1016/j.acclit.2018.03.003

48.

Lev

Radhakrishnan

(2005). The valuation of organization capital. In Corrado

Haltiwanger

Sichel

(Eds.), Measuring capital in the new economy (pp. 73–110). University of Chicago Press.

49.

Lev

Radhakrishnan

Evans

P. C.

(2016). Organizational capital: A CEO’s guide to measuring and managing enterprise intangibles. The Center for Global Enterprise.

50.

Lev

Radhakrishnan

Zhang

(2009). Organization capital. Abacus, 45(3), 275–298. https://doi.org/10.1111/j.1467-6281.2009.00289.x

51.

Qiu

Shen

(2018). Organization capital and mergers and acquisitions. Journal of Financial and Quantitative Analysis, 53(4), 1871–1909. https://doi.org/10.1017/s0022109018000145

52.

Masulis

R. W.

Wang

Xie

(2007). Corporate governance and acquirer returns. The Journal of Finance, 62(4), 1851–1889. https://doi.org/10.1111/j.1540-6261.2007.01259.x

53.

McFadden

(1974). Conditional logit analysis of qualitative choice behavior. In Zarembka

(Ed.), Frontier in econometrics (pp. 105–142). Academic Press.

54.

Miceli

M. P.

Near

J. P.

(1992). Blowing the whistle: The organizational and legal implications for companies and employees. Lexington Books.

55.

Miceli

M. P.

Near

J. P.

Dworkin

T. M.

(2008). Whistle-blowing in organizations. Psychology Press.

56.

Morrison

E. W.

Milliken

F. J.

(2000). Organizational silence: A barrier to change and development in a pluralistic world. Academy of Management Review, 25(4), 706–725. https://doi.org/10.2307/259200

57.

Near

J. P.

Miceli

M. P.

(1985). Organizational dissidence: The case of whistle-blowing. Journal of Business Ethics, 4(1), 1–16. https://doi.org/10.1007/bf00382668

58.

Petersen

M. A.

(2009). Estimating standard errors in finance panel data sets: Comparing approaches. Review of Financial Studies, 22(1), 435–480. https://doi.org/10.1093/rfs/hhn053

59.

Prescott

Visscher

(1980). Organization capital. Journal of Political Economy, 88(3), 446–461. https://doi.org/10.1086/260879

60.

Rice

M. E.

Harris

G. T.

(2005). Comparing effect sizes in follow-up studies: ROC area, Cohen’s d, and r. Law and Human Behavior, 29(5), 615–620. https://doi.org/10.1007/s10979-005-6832-7

61.

Roberts

M. R.

Whited

T. M.

(2013). Endogeneity in empirical corporate finance. In Constantinides

(Ed.), Handbook of the economics of finance (2, pp. 493–572). Elsevier. https://doi.org/10.1016/b978-0-44-453594-8.00007-0

62.

Rupp

D. E.

Bell

C. M.

(2010). Extending the deontic model of justice: Moral self-regulation in third-party responses to injustice. Business Ethics Quarterly, 20(1), 89–106. https://doi.org/10.5840/beq20102017

63.

Sarbanes-Oxley Act of

2002 [SOX], Public law 107-204 [H.R. 3763] (2002). https://www.congress.gov/bill/107th-congress/house-bill/3763/text

64.

Schmidt

(2005). “Whistle blowing” regulation and accounting standards enforcement in Germany and Europe—an economic perspective. International Review of Law and Economics, 25(2), 143–168. https://doi.org/10.1016/j.irle.2005.06.001

65.

Seifert

D. L.

Sweeney

J. T.

Joireman

Thornton

J. M.

(2010). The influence of organizational justice on accountant whistleblowing. Accounting, Organizations and Society, 35(7), 707–717. https://doi.org/10.1016/j.aos.2010.09.002

66.

Shen

C. H.

Zhang

(2013). CEO risk incentives and firm performance following R&D increases. Journal of Banking & Finance, 37(4), 1176–1194. https://doi.org/10.1016/j.jbankfin.2012.11.018

67.

Smith

(2010). The role of whistle-blowing in governing well: Evidence from the Australian public sector. The American Review of Public Administration, 40(6), 704–721. https://doi.org/10.1177/0275074010370201

68.

Soltes

(2020). Paper versus practice: A field investigation of integrity hotlines. Journal of Accounting Research, 58(2), 429–472. https://doi.org/10.1111/1475-679x.12302

69.

Stubben

S. R.

Welch

K. T.

(2020). Evidence on the use and efficacy of internal whistleblowing systems. Journal of Accounting Research, 58(2), 473–518. https://doi.org/10.1111/1475-679x.12303

70.

Trevino

L. K.

Weaver

G. R.

(2001). Organizational justice and ethics program “follow-through”: Influences on employees’ harmful and helpful behavior. Business Ethics Quarterly, 11(4), 651–671. https://doi.org/10.2307/3857765

71.

U.S. Department of Labor Occupational Safety and Health Administration [OSHA] . (2012). Whistleblower investigation manual. Government Printing Office.

72.

Wilde

J. H.

(2017). The deterrent effect of employee whistleblowing on firms’ financial misreporting and tax aggressiveness. The Accounting Review, 92(5), 247–280. https://doi.org/10.2308/accr-51661

73.

Yermack

(1996). Higher market valuation of companies with a small board of directors. Journal of Financial Economics, 40(2), 185–211. https://doi.org/10.1016/0304-405x(95)00844-5

Can Whistleblowing Improve Organizational Effectiveness? Evidence From Financial Reporting Misconduct

Abstract

Background

Purpose

Research Design

Study Sample

Results

Conclusions

Keywords

Introduction

Understanding the Whistleblowing Process

Institutional Background

Whistleblowing Process

Prior Literature and Hypothesis Development

Research Design

Measuring Organizational Capital

Whistleblowing Sample

Propensity Score-Matched Sample

Changes in Organizational Capital

Empirical Results

Changes in Organizational Capital: Main Results

Alternative Measures of Organizational Capital: Robustness Check

Qualitative Measures of Organizational Capital: Robustness Check

Entropy-Balanced Sample: Robustness Check

Additional Analyses

Determinants of Organizational Capital Improvement Decisions

Impact on Future External Whistleblowing Incidence

Press and OSHA Subsamples

Brand Capital Investment

Conclusion

Footnotes

Author’s Note

Declaration of Conflicting Interests

Funding

ORCID iDs

Notes

Appendix

References