This study investigates the relationship between corporate integrity culture and the cost of equity capital.
Using firm-level integrity culture data developed by Li et al. (2021), we find a significant negative association between corporate integrity culture and the cost of equity capital. We estimate an ordinary least squares regression model to investigate the relationship between corporate integrity culture and the implied cost of equity.
Our analysis reveals that corporate integrity culture reduces the cost of equity capital by mitigating the risk of accounting malfeasance. These findings remain after addressing potential endogeneity concerns.
This study makes several important contributions to the literature. First, it is the first to empirically investigate whether corporate integrity culture enhances firm value by reducing the cost of equity capital and establishing a previously unexplored mechanism. Second, this study contributes to the existing literature on factors affecting the cost of equity capital by identifying integrity culture as a key factor and emphasizing the role of accounting malfeasance risk as a channel through which it influences the cost of equity capital. Third, using firm-level integrity culture data, this study addresses gaps in the finance and accounting literature, where integrity culture has received limited attention despite its significance. Using firm-level data enables a more precise and nuanced analysis, overcoming the limitations of traditional proxies such as surveys or regional religious measures. This research broadens the understanding of how corporate integrity culture shapes financial outcomes and provides valuable insights for both academic and practical applications.
Corporate culture and ethics are the foundation of a resilient, high-performing organization. They influence decision-making, risk appetite, and overall business integrity. For any organization, ensuring a strong ethical foundation shouldn’t just be a regulatory requirement, it should be viewed as a strategic advantage [1].
PricewaterhouseCoopers LLP
1. Introduction
Prior studies define corporate integrity culture as the shared values and beliefs related to compliance, trustworthiness and ethics (Altamuro, Gray, & Zhang, 2022). Corporate integrity culture discussions appear in earnings conference calls and online forums such as Glassdoor and companies' websites [2]. For instance, the Fortune 500 company Salesforce's corporate website clearly articulates its values, including trust, customer success, innovation and equality. On its website, Salesforce states: “We earn the trust of our customers, employees and extended family through transparency, security, compliance, privacy and performance. And we deliver the industry's most trusted infrastructure [3]. Prior research in accounting and finance has explored how corporate culture integrity influences business operations, including operational compliance (Altamuro et al., 2022), earnings manipulation (Jia, 2013) and internal control practices (Shu, Chen, & Lin, 2018). Recent research has identified a link between corporate integrity culture and firm value (Li et al., 2021; Graham, Harvey, Popadak, & Rajgopal, 2022). However, there is limited empirical evidence on the mechanism through which integrity culture affects firm value. In this paper, we hypothesize that corporate integrity culture influences firm value by reducing the cost of equity capital [4].
Executives consider corporate culture one of the least researched factors contributing to firm value (Graham et al., 2022). After reviewing various studies on corporate culture, Grennan and Li (2023) emphasize the need for further research on its implications. Similarly, Graham et al. (2022) call for future studies to investigate which dimensions of corporate culture drive firm value. One key dimension of corporate culture is integrity culture, yet how it impacts firm value remains unclear, largely due to challenges in measuring corporate integrity culture at the firm level. Prior research has primarily relied on survey data (Graham et al., 2022), manager-level integrity data (Kizirian, Mayhew, & Sneathen, 2005), or proxies such as county-level religious data (McGuire, Omer, & Sharp, 2012) to proxy for corporate integrity culture. This study uses the newly available firm-level integrity culture data derived from the transcripts of the Question and Answer section of the earnings conference calls (Li et al., 2021) [5].
Our analysis aims to explore whether corporate integrity culture affects a firm's cost of equity capital, using the implied cost of equity capital (ICC) as the dependent variable. We find a consistently negative and statistically significant relation between corporate integrity culture and the cost of equity capital. Economically, a one-unit increase in corporate integrity culture corresponds to a 5.0 basis point reduction in the cost of equity capital and a one-standard-deviation increase leads to a 4.44 basis point decrease. Overall, the findings confirm that firms with stronger corporate integrity cultures experience lower costs of equity capital [6].
We then validate the results by providing evidence that the reduction in the risk of accounting malfeasance due to a strong integrity culture explains the decrease in the cost of equity capital. This explanation is grounded in prior research that has shown a clear relation between integrity culture and accounting malfeasance. Specifically, Li et al. (2021) suggest that a strong culture of integrity is negatively associated with the probability of financial restatement (restatement afterward), a key indicator of accounting malfeasance. Prior research also shows that a reduction in restatements directly lowers a firm's cost of equity capital (Hribar & Jenkins, 2004). Our results corroborate these findings and further emphasize that the reduction in accounting malfeasance risk, driven by a strong integrity culture, serves as a key channel through which firms experience a lower cost of equity capital.
Our cross-sectional analysis further examines the relation between integrity and the cost of equity capital by emphasizing that integrity is more critical in low-trust environments. Trust is operationalized using state-level trust culture data and county-level religiosity data. The findings show that the negative relationship between integrity and the cost of equity capital is significant only in areas with lower trust or religiosity levels, which suggests that firms in such regions benefit more from a strong culture of integrity through lower equity financing costs.
Our model may have endogeneity issues. Therefore, we also assess the robustness of our main findings by controlling for endogeneity through a two-stage least squares (2SLS) approach, a propensity score matching approach (PSM) and a lead-lag approach. First, we adopt a 2SLS approach using the geographic distance between a firm's headquarters and its closest religious site as an instrument. We find that this distance is negatively related to corporate integrity culture and the predicted corporate integrity culture measure is negatively related to a firm's cost of equity capital. Second, to ensure that firms with strong and weak corporate integrity are not systematically different, we use a PSM approach. We find that our results remain robust for a reduced sample of PSM-strong corporate integrity firms and PSM-control firms. Third, to address concerns about reverse causality, we examine whether the corporate integrity culture of the previous year influences the cost of equity capital of the current year by using lagged corporate integrity culture and lagged control variables as regressors. We find that the corporate integrity culture of the previous year is negatively related to the implied cost of equity capital of the current year, which suggests that corporate integrity culture change precedes the change in the cost of equity capital. These findings indicate a causal impact of corporate integrity culture on the cost of equity capital.
Our study contributes to the literature in several ways. Prior research highlights a connection between corporate culture and firm value, but no studies test whether culture can increase firm value by reducing the cost of equity capital. To the best of our knowledge, this paper is the first to investigate the effect of corporate integrity culture on the cost of equity capital, which is a potential mechanism through which integrity culture can impact firm value.
Second, we add to the literature on the determinants of the cost of equity capital. One stream of literature focuses on the effect of information asymmetry on the cost of equity capital (Botosan, 1997; Botosan & Plumlee, 2002; Botosan, Plumlee, & Xie, 2004; Francis, Nanda, & Olsson, 2008). Another stream of literature studies how various risks affect the cost of equity capital, such as customer concentration risk (Dhaliwal, Judd, Serfling, & Shaikh, 2016), company reputation risk (Cao, Myers, & Omer, 2015), social capital risk (Richardson & Welker, 2001) and corporate social responsibility risk (Dhaliwal, Li, Tsang, & Yang, 2011; El Ghoul et al., 2011). Our research adds to these studies, as it specifically focuses on the risk of accounting malfeasance.
Third, we add to the literature on the importance of corporate integrity culture. Previous studies have found that corporate integrity culture can influence the success of a business and affect the satisfaction and commitment of employees (Li et al., 2021; Graham et al., 2022). However, the finance and accounting literature has paid limited attention to corporate integrity culture, despite its proven significance. This is in part because the concept of integrity culture is somewhat abstract. In this study, we use firm-level integrity data developed by Li et al. (2021). The firm-level integrity culture data allow for a more precise and comprehensive analysis of how integrity culture influences investor perceptions and financial outcomes, thus pushing the boundaries of traditional approaches that rely on proxies like survey data or regional religious measures.
2. Background and hypothesis development
2.1 Background of corporate integrity culture
Corporate integrity culture refers to the values, norms and ethical standards that govern behavior within an organization (Altamuro et al., 2022). It emphasizes honesty, transparency and accountability, fostering an environment where ethical behavior is prioritized over short-term gains (Li et al., 2021). In recent decades, high-profile corporate scandals such as Enron, WorldCom and Lehman Brothers have fueled academic and practical interest in corporate integrity. For instance, the US Congress responded by enacting legislation such as the Sarbanes (2002), imposing stringent corporate governance requirements. These laws aim to create transparency and accountability, ensuring that organizations maintain integrity in financial reporting, decision-making and operations.
Integrity in financial reporting is critical for several reasons, as it directly impacts the trust and long-term success of an organization. First, integrity builds trust with key stakeholders such as investors, customers, employees and regulators (Paine, 1994). When financial reporting and decision-making processes are transparent and ethical, stakeholders can rely on the accuracy and honesty of the information provided (Bayou, Reinstein, & Williams, 2011). This trust is fundamental for maintaining credibility in the marketplace and ensuring ongoing support from investors (Garrett, Hoitash, & Prawitt, 2014). Without integrity, the organization's reputation is at risk, which can lead to a loss of confidence. Second, a strong culture of integrity helps prevent fraud, financial manipulation and unethical behavior (Li et al., 2021). High-profile corporate scandals, such as Enron and WorldCom, stemmed from a lack of integrity in financial reporting and decision-making, leading to devastating impacts on employees and shareholders. Integrity creates safeguards against such malpractices, ensuring accurate financial records and ethical operational practices.
Due to the lack of corporate integrity data, prior studies have generally used various non-firm-level measures as proxies for integrity. For instance, DeBacker, Heim, and Tran (2015) find that corruption culture increases corporate tax evasion in the USA. Biggerstaff, Cicero, and Puckett (2015) find that firms with CEOs who backdate options are more likely to have an unethical culture and overstate earnings. Hilary and Hui (2009) find that firms located in counties with higher levels of religiosity have lower return volatility. Guiso, Sapienza, and Zingales (2015) find that, when employees perceive top managers as trustworthy and ethical, the firm's performance is stronger and they further suggest that conducting a field experiment is the most convincing approach to effectively demonstrate the causal relationship between integrity and firm performance. Answering the call of Guiso et al. (2015), Graham et al. (2022) conducted the first field study about corporate culture [7].
Recently, researchers have begun to quantify corporate integrity culture on a large scale with the availability of large earnings conference call data and the development of machine learning methods. According to Graham et al. (2022), earnings conference calls are considered the most effective tool for assessing a company's corporate culture. Therefore, drawing on transcripts from earnings calls, Li et al. (2021) employ the latest machine learning method, the word-embedding model, to assess the manifestation of the five core corporate cultural values – innovation, integrity, quality, respect and teamwork – in their study [8]. They find that corporate integrity culture is influenced by top management, unlike national cultural values that are deeply ingrained. Therefore, they assess corporate integrity culture through earnings calls held by managers such as CEOs. Further, as opposed to a company's website or press releases (as in Guiso et al., 2015), which might not offer opportunities for “empty promises” regarding advertised values, earnings calls are not primarily focused on discussing a company's values but rather its business operations and financial performance (Li et al., 2021). Therefore, this study uses the corporate integrity culture measure developed by Li et al. (2021) [9].
2.2 Hypothesis development
A strong integrity culture embeds ethical principles such as honesty, transparency and accountability throughout the organization, encouraging ethical decision-making at all levels. This ethical framework helps prevent financial misreporting and fraudulent activities, thereby reducing the risk of accounting malfeasance (Li et al., 2021).
One of the main reasons why corporate integrity culture reduces the likelihood of restatement is by fostering strong internal governance mechanisms and ethical decision-making. When ethical conduct is prioritized, employees and managers are more likely to report misconduct or questionable accounting practices before they escalate into significant issues (Rose et al., 2021). Further, internal controls are also more robust in organizations with high integrity, as employees are aligned with ethical goals and accountability (Kizirian et al., 2005). These enhancements in governance mechanisms help mitigate the risk of both intentional and unintentional financial misreporting.
Accounting restatements erode stakeholder confidence, as they indicate poor governance, weak internal controls, or management dishonesty (Cahan, Chen, & Chen, 2024) [10]. Restatements also trigger negative market reactions, increased scrutiny from regulatory bodies and legal penalties, all of which increase the firm's risk profile and result in higher costs of capital (Fragoso, Peixinho, Coelho, & Paiva, 2020). Furthermore, restatements lead to reputational damage, which can diminish a company's credibility and its ability to attract new investors (Cao et al., 2015). In contrast, firms that promote a culture of integrity are more likely to ensure accurate financial reporting, adhering to ethical guidelines and regulatory standards, thus avoiding these costly outcomes (Graham et al., 2022). Consistent with the above views, prior literature finds that a reduction in restatements and fraud directly affects a firm's cost of equity capital (Hribar & Jenkins, 2004). Therefore, a strong corporate integrity culture not only improves financial reporting accuracy but also enhances the firm's overall risk profile. As a result, such firms are able to attract investors at a lower cost of equity capital. Thus, we propose the following hypothesis:
Firms with a strong corporate integrity culture have a lower cost of equity capital.
3. Data and empirical methodology
3.1 Sample selection
Our primary dataset is created by combining data from multiple sources, including the corporate integrity culture data from Li et al. (2021), financial statement data from Compustat, stock return data from the Center for Research in Security Prices files and earnings forecast data from the Institutional Brokers' Estimate System [11]. The intersection of these databases forms the basis of our analysis. Our primary dataset consists of 17,910 US firm-year observations covering the period from 2002 to 2020 [12]. We exclude utilities (SIC 4900-4999) and financial firms (SIC 6000-6999) from our analysis. To minimize the impact of outliers, we winsorize all continuous variables at their 1st and 99th percentiles.
Panel A of Table 1 displays the summary statistics for the implied cost of equity estimates. Panel B shows the summary statistics for the independent variable and the other control variables. The definitions of all variables are provided in Appendix B.
Summary statistics
| Mean | SD | P25 | Median | P75 | N | |
|---|---|---|---|---|---|---|
| Panel A. summary statistics for implied cost of equity capital (%) | ||||||
| Cost of equity | 9.671 | 2.547 | 8.045 | 9.340 | 10.953 | 17,910 |
| Cost of equity (GLS) | 9.880 | 2.922 | 8.073 | 9.747 | 11.523 | 17,910 |
| Cost of equity (CT) | 8.000 | 2.799 | 6.238 | 7.801 | 9.431 | 17,910 |
| Cost of equity (MPEG) | 10.723 | 3.893 | 8.209 | 9.828 | 12.365 | 17,910 |
| Cost of equity (OJ) | 10.058 | 2.747 | 8.323 | 9.640 | 11.353 | 17,910 |
| Panel B. summary statistics for control variables | ||||||
| Corporate Integrity | 2.096 | 0.888 | 1.463 | 1.943 | 2.578 | 17,910 |
| Beta Value-Weighted | 1.232 | 0.549 | 0.854 | 1.159 | 1.536 | 17,910 |
| Idiosyncratic Risk | 0.426 | 0.192 | 0.285 | 0.389 | 0.523 | 17,910 |
| Market Value of Equity | 7.735 | 1.643 | 6.521 | 7.586 | 8.797 | 17,910 |
| Market-to-Book | 4.010 | 4.821 | 1.714 | 2.653 | 4.284 | 17,910 |
| Book Leverage | 0.222 | 0.181 | 0.050 | 0.211 | 0.345 | 17,910 |
| Momentum | 0.152 | 0.428 | −0.107 | 0.096 | 0.322 | 17,910 |
| Analyst Forecast Dispersion | 0.111 | 0.263 | 0.017 | 0.032 | 0.078 | 17,910 |
| Long-Term Growth Rate | 0.151 | 0.090 | 0.100 | 0.140 | 0.188 | 17,910 |
| Return on Assets | 0.053 | 0.071 | 0.023 | 0.054 | 0.090 | 17,910 |
| Mean | SD | P25 | Median | P75 | N | |
|---|---|---|---|---|---|---|
| Panel A. summary statistics for implied cost of equity capital (%) | ||||||
| Cost of equity | 9.671 | 2.547 | 8.045 | 9.340 | 10.953 | 17,910 |
| Cost of equity (GLS) | 9.880 | 2.922 | 8.073 | 9.747 | 11.523 | 17,910 |
| Cost of equity (CT) | 8.000 | 2.799 | 6.238 | 7.801 | 9.431 | 17,910 |
| Cost of equity (MPEG) | 10.723 | 3.893 | 8.209 | 9.828 | 12.365 | 17,910 |
| Cost of equity (OJ) | 10.058 | 2.747 | 8.323 | 9.640 | 11.353 | 17,910 |
| Panel B. summary statistics for control variables | ||||||
| Corporate Integrity | 2.096 | 0.888 | 1.463 | 1.943 | 2.578 | 17,910 |
| Beta Value-Weighted | 1.232 | 0.549 | 0.854 | 1.159 | 1.536 | 17,910 |
| Idiosyncratic Risk | 0.426 | 0.192 | 0.285 | 0.389 | 0.523 | 17,910 |
| Market Value of Equity | 7.735 | 1.643 | 6.521 | 7.586 | 8.797 | 17,910 |
| Market-to-Book | 4.010 | 4.821 | 1.714 | 2.653 | 4.284 | 17,910 |
| Book Leverage | 0.222 | 0.181 | 0.050 | 0.211 | 0.345 | 17,910 |
| Momentum | 0.152 | 0.428 | −0.107 | 0.096 | 0.322 | 17,910 |
| Analyst Forecast Dispersion | 0.111 | 0.263 | 0.017 | 0.032 | 0.078 | 17,910 |
| Long-Term Growth Rate | 0.151 | 0.090 | 0.100 | 0.140 | 0.188 | 17,910 |
| Return on Assets | 0.053 | 0.071 | 0.023 | 0.054 | 0.090 | 17,910 |
Note(s): The table presents the descriptive statistics for the variables used in the regressions. The sample consists of publicly traded non-financial and non-utility firms in the United States. The full sample includes 17,910 firm-year observations from 2002 to 2020. Panel A presents summary statistics for implied cost of equity estimates. Panel B presents summary statistics for control variables. All continuous variables are winsorized at their 1st and 99th percentiles. All variables are defined in Appendix A
3.2 Measures of the cost of equity capital
Using the four cost of equity models introduced by Claus and Thomas (2001), Gebhardt, Lee, and Swaminathan (2001), Easton (2004) and Ohlson and Juettner-Nauroth (2005), we estimate the cost of equity capital implied in current stock prices and analysts' earnings forecasts. The first two models are based on Ohlson's (1995) residual income valuation model, while the latter two models are based on the abnormal earnings growth valuation model developed by Ohlson and Juettner-Nauroth (2005). A detailed description of the cost of equity capital estimates is provided in Appendix C. Due to the lack of consensus in the literature on which models perform best or how they should be evaluated, we follow prior literature and use the mean of the estimates from the four models as our measure of the cost of equity capital to alleviate the effect of measurement errors associated with any one particular model (Hail & Leuz, 2006; Dhaliwal et al., 2016) [13]. In general, our estimates of the cost of equity capital are similar to those found in previous studies (e.g., Dhaliwal et al., 2016) [14].
3.3 Measures of corporate integrity culture
We measure corporate integrity culture using two approaches: the integrity culture score from Li et al. (2021) and a binary indicator for whether a firm exhibits a strong integrity culture. Consistent with Li et al. (2021), we categorize firms into two groups: those with a strong integrity culture and those with a weak one. The key independent variable, the Corporate Integrity Dummy, is a binary indicator set to one if a firm's integrity value falls within the top quintile, and zero otherwise.
Li et al.’s (2021) word-embedding methodology offers a sophisticated means of capturing integrity culture by analyzing authentic managerial discourse in the spontaneous Q&A segments of earnings calls. This measure reflects integrity as expressed through discourse rather than the organization's broader culture. It may, for example, detect conversations about building stakeholder relationships, yet overlook expressions of integrity culture embedded in formal policies or routine decisions that do not appear in earnings calls. Therefore, we argue that this measure provides a meaningful proxy for a firm's integrity culture, though it may not constitute a comprehensive evaluation of a firm's overall integrity culture [15].
3.4 Research design
We estimate the following ordinary least squares regression model to investigate the relation between corporate integrity culture and the implied cost of equity:
where ICC is our measure of the estimated implied cost of equity [16]. Corporate Integrity Culture is our variable of interest. We include control variables such as equity beta (value-weighted), idiosyncratic risk, market value of equity, market-to-book ratio, book leverage, momentum, analyst forecast dispersion, long-term growth rate and return on assets, following Dhaliwal et al. (2016). We also include year-fixed effects and firm-fixed effects when estimating a variety of regressions. Test statistics are based on standard errors adjusted for heteroscedasticity and clustered at the firm level.
4. Empirical results
4.1 Corporate integrity culture and the cost of equity capital
We initiate our analysis by investigating whether a strong corporate integrity culture impacts a firm's cost of equity. Table 2 shows the results of this analysis. In Columns (1) to (4), the dependent variable is the implied cost of equity. Column (1) reports the results from the main regression model, including corporate integrity culture score (Corporate Integrity), Market Value of Equity, Market-to-Book, Book Leverage, Momentum, Analyst Forecast Dispersion, Long-Term Growth Rate, Return on Assets, year-fixed effects and firm-fixed effects [17]. The results indicate a negative and statistically significant correlation between corporate integrity culture and a firm's cost of equity. Column (2) reports the results from the model including all control variables, including equity beta and idiosyncratic risk [18]. The coefficient of corporate integrity culture remains significantly negative after controlling for the known determinants of the cost of equity capital. In economic terms, the coefficient estimate suggests that a unit increase in corporate integrity culture decreases the cost of equity capital by 5.0 basis points. For a one-standard-deviation increase in Corporate Integrity, the economic impact is 4.44 basis points decrease in the cost of equity capital [19]. Similar to prior studies (Ashbaugh-Skaife, Collins, Kinney, & Lafond, 2009; Dhaliwal et al., 2016), the cost of equity capital increases in both beta and idiosyncratic risk. We obtain similar results when the independent variable is a Corporate Integrity Dummy [20]. In addition, Table 2 Columns (1) to (4) reveal that the signs of the estimated coefficients on the control variables agree with the prior research findings presented in the literature. For instance, there is a positive correlation between the implied cost of equity and equity beta, idiosyncratic risk, book leverage and the long-term growth rate. Moreover, firm size, market-to-book ratio, price momentum and profitability exhibit a negative correlation with the cost of equity capital. In summary, the findings presented in Table 2 support our notion that a firm with a strong corporate integrity culture experiences a lower cost of equity.
Corporate integrity culture and the implied cost of equity capital
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity | −0.047** | −0.050** | ||
| (−2.162) | (−2.321) | |||
| Corporate Integrity Dummy | −0.106*** | −0.110*** | ||
| (−2.583) | (−2.664) | |||
| Beta Value-Weighted | 0.081** | 0.081** | ||
| (2.244) | (2.247) | |||
| Idiosyncratic Risk | 0.799*** | 0.795*** | ||
| (5.086) | (5.070) | |||
| Market Value of Equity | −0.451*** | −0.399*** | −0.451*** | −0.400*** |
| (−8.885) | (−7.751) | (−8.903) | (−7.770) | |
| Market-to-Book | −0.030*** | −0.031*** | −0.030*** | −0.031*** |
| (−4.904) | (−5.115) | (−4.916) | (−5.126) | |
| Book Leverage | 1.755*** | 1.713*** | 1.752*** | 1.711*** |
| (8.064) | (7.920) | (8.057) | (7.911) | |
| Momentum | −0.270*** | −0.309*** | −0.269*** | −0.309*** |
| (−6.120) | (−6.961) | (−6.121) | (−6.960) | |
| Analyst Forecast Dispersion | −0.028 | −0.052 | −0.026 | −0.050 |
| (−0.355) | (−0.663) | (−0.332) | (−0.638) | |
| Long-Term Growth Rate | 1.783*** | 1.724*** | 1.787*** | 1.729*** |
| (6.363) | (6.184) | (6.386) | (6.211) | |
| Return on Assets | −1.768*** | −1.569*** | −1.766*** | −1.568*** |
| (−4.876) | (−4.342) | (−4.871) | (−4.337) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 17,910 | 17,910 | 17,910 | 17,910 |
| Adjusted R2 | 0.628 | 0.629 | 0.628 | 0.629 |
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity | −0.047** | −0.050** | ||
| (−2.162) | (−2.321) | |||
| Corporate Integrity Dummy | −0.106*** | −0.110*** | ||
| (−2.583) | (−2.664) | |||
| Beta Value-Weighted | 0.081** | 0.081** | ||
| (2.244) | (2.247) | |||
| Idiosyncratic Risk | 0.799*** | 0.795*** | ||
| (5.086) | (5.070) | |||
| Market Value of Equity | −0.451*** | −0.399*** | −0.451*** | −0.400*** |
| (−8.885) | (−7.751) | (−8.903) | (−7.770) | |
| Market-to-Book | −0.030*** | −0.031*** | −0.030*** | −0.031*** |
| (−4.904) | (−5.115) | (−4.916) | (−5.126) | |
| Book Leverage | 1.755*** | 1.713*** | 1.752*** | 1.711*** |
| (8.064) | (7.920) | (8.057) | (7.911) | |
| Momentum | −0.270*** | −0.309*** | −0.269*** | −0.309*** |
| (−6.120) | (−6.961) | (−6.121) | (−6.960) | |
| Analyst Forecast Dispersion | −0.028 | −0.052 | −0.026 | −0.050 |
| (−0.355) | (−0.663) | (−0.332) | (−0.638) | |
| Long-Term Growth Rate | 1.783*** | 1.724*** | 1.787*** | 1.729*** |
| (6.363) | (6.184) | (6.386) | (6.211) | |
| Return on Assets | −1.768*** | −1.569*** | −1.766*** | −1.568*** |
| (−4.876) | (−4.342) | (−4.871) | (−4.337) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 17,910 | 17,910 | 17,910 | 17,910 |
| Adjusted R2 | 0.628 | 0.629 | 0.628 | 0.629 |
Note(s): The table presents the relation between corporate integrity culture and the implied cost of equity capital. The dependent variable is the implied cost of equity capital. The independent variables of interest are Corporate Integrity in Columns (1) and (2) and Corporate Integrity Dummy in Columns (3) and (4). Other variables are defined in Appendix A. The regressions also include year fixed effects and firm fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
4.2 Channel analysis
4.2.1 Relation between corporate integrity culture and accounting malfeasance
Li et al. (2021) provide evidence that the cultural value of integrity is negatively associated with accounting malfeasance, specifically focusing on financial restatements [21]. This finding suggests that firms with a strong integrity culture are less likely to engage in practices that lead to accounting restatements, which are often indicative of errors or even fraudulent reporting. In this section, we further validate the relation between corporate integrity culture and accounting restatements.
A strong culture of integrity fosters ethical principles such as honesty, transparency and accountability, embedding these values throughout the organization and encouraging ethical decision-making at all levels (Li et al., 2021). By prioritizing ethical conduct, firms can prevent fraud, financial manipulation and unethical behavior, thereby reducing the risk of corporate malfeasance (Murphy & Dacin, 2011). Consequently, firms with a strong corporate integrity culture are less likely to experience financial restatements, as transparency and adherence to reporting standards are reinforced. Table 3 examines the relation between corporate integrity and financial restatements using a logistic model. The dependent variable, Restatement (based on Hennes, Leone, & Miller, 2008), is our measure of the risk of corporate malfeasance. Corporate Integrity is the primary independent variable of interest. A negative coefficient indicates that higher corporate integrity is associated with a lower likelihood of financial restatements. In all regression models (Table 3 Columns (1) and (2)), Corporate Integrity shows a negative and statistically significant coefficient. This relation remains even after controlling for various firm characteristics and incorporating year-fixed effects and firm-fixed effects. Our results still hold when we use Corporate Integrity Dummy (Column (3) and (4)). The findings support the notion that a strong corporate integrity culture contributes to better financial reporting practices, thereby reducing the likelihood of accounting malfeasance (Li et al., 2021).
Corporate integrity culture and financial reporting restatement
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity | −0.087** | −0.088** | ||
| (−2.302) | (−2.329) | |||
| Corporate Integrity Dummy | −0.208** | −0.217*** | ||
| (−2.575) | (−2.673) | |||
| Market Value of Equity | −0.075** | −0.077*** | ||
| (−2.537) | (−2.586) | |||
| Book Leverage | 0.529* | 0.530* | ||
| (1.880) | (1.881) | |||
| Return on Assets | −2.534*** | −2.546*** | ||
| (−3.447) | (−3.469) | |||
| Loss | 0.018 | 0.016 | ||
| (0.155) | (0.140) | |||
| Market-to-Book | −0.018 | −0.018 | ||
| (−1.589) | (−1.599) | |||
| Restructuring Charge | 0.136* | 0.134* | ||
| (1.773) | (1.759) | |||
| Special Items | 3.039*** | 3.071*** | ||
| (2.687) | (2.723) | |||
| M&A | −0.046 | −0.044 | ||
| (−0.677) | (−0.661) | |||
| Foreign transaction | −0.050 | −0.050 | ||
| (−0.496) | (−0.500) | |||
| New issuance | 0.136 | 0.136 | ||
| (1.125) | (1.128) | |||
| Firm age | −0.494 | −0.488 | ||
| (−0.870) | (−0.864) | |||
| Litigiousness | −0.047 | −0.047 | ||
| (−0.718) | (−0.730) | |||
| Capital intensity | −0.639* | −0.641* | ||
| (−1.848) | (−1.859) | |||
| Intangible assets | −0.463 | −0.445 | ||
| (−0.670) | (−0.645) | |||
| Momentum | −0.040 | −0.040 | ||
| (−0.596) | (−0.592) | |||
| Interest coverage | −0.102 | −0.228 | ||
| (−0.154) | (−0.350) | |||
| Year fixed effects | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes |
| Observations | 17,280 | 17,280 | 17,280 | 17,280 |
| Adjusted R2 | 0.080 | 0.088 | 0.081 | 0.088 |
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity | −0.087** | −0.088** | ||
| (−2.302) | (−2.329) | |||
| Corporate Integrity Dummy | −0.208** | −0.217*** | ||
| (−2.575) | (−2.673) | |||
| Market Value of Equity | −0.075** | −0.077*** | ||
| (−2.537) | (−2.586) | |||
| Book Leverage | 0.529* | 0.530* | ||
| (1.880) | (1.881) | |||
| Return on Assets | −2.534*** | −2.546*** | ||
| (−3.447) | (−3.469) | |||
| Loss | 0.018 | 0.016 | ||
| (0.155) | (0.140) | |||
| Market-to-Book | −0.018 | −0.018 | ||
| (−1.589) | (−1.599) | |||
| Restructuring Charge | 0.136* | 0.134* | ||
| (1.773) | (1.759) | |||
| Special Items | 3.039*** | 3.071*** | ||
| (2.687) | (2.723) | |||
| M&A | −0.046 | −0.044 | ||
| (−0.677) | (−0.661) | |||
| Foreign transaction | −0.050 | −0.050 | ||
| (−0.496) | (−0.500) | |||
| New issuance | 0.136 | 0.136 | ||
| (1.125) | (1.128) | |||
| Firm age | −0.494 | −0.488 | ||
| (−0.870) | (−0.864) | |||
| Litigiousness | −0.047 | −0.047 | ||
| (−0.718) | (−0.730) | |||
| Capital intensity | −0.639* | −0.641* | ||
| (−1.848) | (−1.859) | |||
| Intangible assets | −0.463 | −0.445 | ||
| (−0.670) | (−0.645) | |||
| Momentum | −0.040 | −0.040 | ||
| (−0.596) | (−0.592) | |||
| Interest coverage | −0.102 | −0.228 | ||
| (−0.154) | (−0.350) | |||
| Year fixed effects | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes |
| Observations | 17,280 | 17,280 | 17,280 | 17,280 |
| Adjusted R2 | 0.080 | 0.088 | 0.081 | 0.088 |
Note(s): The table presents the relation between corporate integrity culture and financial reporting restatement. The dependent variable is accounting restatement (Hennes et al., 2008). The independent variables of interest are Corporate Integrity in Columns (1) and (2) and Corporate Integrity Dummy in Columns (3) and (4). The independent variables of the logistic regression also include firm size (Market Value of Equity), financial leverage (Book Leverage), profitability (Return on Assets), accounting loss (Loss), market-to-book ratio (Market-to-Book), restructuring (Restructuring Charge), special items (Special Items), mergers and acquisitions indicator (M&A), foreign transaction indicator (Foreign Transaction), equity issuance indicator (New Issuance), the log of firm age (Firm Age), litigation dummy (Litigiousness), assets scaled intangible assets (Intangible Assets), capital intensity (Capital Intensity) and interest coverage ratio (Interest Coverage) based upon prior research (Hennes et al., 2008). Other variables are defined in Appendix A. The regressions also include year fixed effects and firm fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
4.2.2 Relation between accounting malfeasance and the cost of equity capital
Accounting restatements undermine stakeholder confidence by signaling poor governance, weak internal controls, or potential management dishonesty (Cahan et al., 2024). They often result in negative market reactions, heightened regulatory scrutiny and legal penalties, all of which elevate the firm's risk profile and increase its cost of capital (Fragoso et al., 2020). Consistent with this view, prior research shows that reducing restatements and fraud directly lowers a firm's cost of equity capital (Hribar & Jenkins, 2004). Therefore, we empirically confirm this finding without hypothesizing the relation between restatement and the cost of equity capital. Table 4 examines the relation between Accounting Restatement and the implied cost of equity capital. Table 4 provides clear evidence that financial restatements are associated with an increase in the implied cost of equity capital.
Financial reporting restatement and the implied cost of equity capital
| (1) | (2) | |
|---|---|---|
| Accounting Restatement | 0.110** | 0.101* |
| (2.063) | (1.890) | |
| Beta Value-Weighted | 0.080** | |
| (2.204) | ||
| Idiosyncratic Risk | 0.786*** | |
| (5.009) | ||
| Market Value of Equity | −0.451*** | −0.400*** |
| (−8.884) | (−7.758) | |
| Market-to-Book | −0.030*** | −0.031*** |
| (−4.924) | (−5.133) | |
| Book Leverage | 1.742*** | 1.702*** |
| (7.998) | (7.860) | |
| Momentum | −0.267*** | −0.306*** |
| (−6.057) | (−6.880) | |
| Analyst Forecast Dispersion | −0.026 | −0.049 |
| (−0.332) | (−0.635) | |
| Long-Term Growth Rate | 1.798*** | 1.740*** |
| (6.410) | (6.238) | |
| Return on Assets | −1.750*** | −1.554*** |
| (−4.824) | (−4.297) | |
| Year Fixed Effects | Yes | Yes |
| Firm Fixed Effects | Yes | Yes |
| Observations | 17,910 | 17,910 |
| Adjusted R2 | 0.628 | 0.629 |
| (1) | (2) | |
|---|---|---|
| Accounting Restatement | 0.110** | 0.101* |
| (2.063) | (1.890) | |
| Beta Value-Weighted | 0.080** | |
| (2.204) | ||
| Idiosyncratic Risk | 0.786*** | |
| (5.009) | ||
| Market Value of Equity | −0.451*** | −0.400*** |
| (−8.884) | (−7.758) | |
| Market-to-Book | −0.030*** | −0.031*** |
| (−4.924) | (−5.133) | |
| Book Leverage | 1.742*** | 1.702*** |
| (7.998) | (7.860) | |
| Momentum | −0.267*** | −0.306*** |
| (−6.057) | (−6.880) | |
| Analyst Forecast Dispersion | −0.026 | −0.049 |
| (−0.332) | (−0.635) | |
| Long-Term Growth Rate | 1.798*** | 1.740*** |
| (6.410) | (6.238) | |
| Return on Assets | −1.750*** | −1.554*** |
| (−4.824) | (−4.297) | |
| Year Fixed Effects | Yes | Yes |
| Firm Fixed Effects | Yes | Yes |
| Observations | 17,910 | 17,910 |
| Adjusted R2 | 0.628 | 0.629 |
Note(s): The table presents the relation between financial reporting statement and the implied cost of equity capital. The dependent variable is the implied cost of equity capital. The independent variable of interest is Accounting Restatement, which indicates a firm has a restatement for a given year. Other variables are defined in Appendix A. The regressions also include year fixed effects and firm fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
4.3 Cross-sectional analysis
Integrity is connected to the trust that exists between individuals, as it forms the foundation of reliable and honest relationships. Therefore, we posit that integrity is more important when trust is low. A company's trustworthiness can impact the relationship between integrity culture and the cost of equity capital for several reasons. First, when a company is viewed as trustworthy, investors are more likely to have confidence in its ability to deliver strong financial performance over time. This increased confidence can lead to a reduction in the cost of equity capital, as investors may be willing to accept a lower rate of return on their investment. Second, trustworthy companies are generally viewed as lower-risk investments. This perception of lower risk can result in a lower cost of equity capital, as investors may be willing to accept a lower rate of return in exchange for a lower level of risk (Phillips & Zhdanov, 2013). Third, companies that are viewed as trustworthy are often more likely to receive favorable terms from lenders and investors. This improved access to capital can help lower the cost of equity capital (El Ghoul et al., 2011).
We operationalize trust using local state-level trust culture data from Hayes, Jiang, and Pan (2021) and county-level religious adherent data from the Churches and Church Membership files of the Association of Religion Data Archives. Mazar, Amir, and Ariely (2008) show that participants who were reminded of their religious beliefs before taking a test were less likely to cheat compared to those who were not reminded. To investigate the effect of trust on the relationship between integrity culture and the cost of equity capital, we perform a cross-sectional analysis by dividing the full sample into two subsamples: firms in high-trust environments and firms in low-trust environments. First, to perform a cross-sectional analysis, we divide our sample based on the median of the state-level trust index. We find that the negative relation between integrity and the cost of equity capital is significant only for the subsample with a lower trust index (see Table 5 Columns (1) and (2)). Similarly, we also find that the negative relation between integrity and the cost of equity capital is significant only for the low religiosity subsample (see Table 5 Columns (3) and (4)). To summarize, our cross-sectional analysis reveals that the benefits of a strong integrity culture are most evident in low-trust and low-religiosity environments, where the capital market places a premium on firms that cultivate integrity.
Corporate integrity culture and the implied cost of equity capital: cross-sectional analysis
| Trust index | Religiosity | |||
|---|---|---|---|---|
| Low | High | Low | High | |
| (1) | (2) | (3) | (4) | |
| Corporate Integrity | −0.061** | −0.038 | −0.067** | −0.031 |
| (−2.059) | (−1.232) | (−2.227) | (−1.015) | |
| Beta Value-Weighted | 0.091* | 0.073 | 0.164*** | 0.011 |
| (1.710) | (1.497) | (2.959) | (0.238) | |
| Idiosyncratic Risk | 1.278*** | 0.304 | 1.014*** | 0.627*** |
| (5.628) | (1.439) | (4.504) | (2.878) | |
| Market Value of Equity | −0.434*** | −0.357*** | −0.318*** | −0.466*** |
| (−5.579) | (−5.163) | (−4.064) | (−6.861) | |
| Market-to-Book | −0.026*** | −0.034*** | −0.032*** | −0.030*** |
| (−3.066) | (−4.020) | (−3.882) | (−3.403) | |
| Book Leverage | 1.376*** | 2.013*** | 1.442*** | 1.941*** |
| (4.141) | (7.045) | (4.348) | (6.756) | |
| Momentum | −0.361*** | −0.258*** | −0.420*** | −0.215*** |
| (−5.595) | (−4.216) | (−6.400) | (−3.630) | |
| Analyst Forecast Dispersion | −0.111 | 0.008 | −0.162 | 0.033 |
| (−0.953) | (0.076) | (−1.303) | (0.341) | |
| Long-Term Growth Rate | 2.059*** | 1.376*** | 2.496*** | 0.948*** |
| (5.194) | (3.502) | (5.937) | (2.625) | |
| Return on Assets | −1.670*** | −1.497*** | −1.854*** | −1.362*** |
| (−3.126) | (−3.006) | (−3.181) | (−2.967) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 9,099 | 8,811 | 8,854 | 9,056 |
| Adjusted R2 | 0.614 | 0.646 | 0.618 | 0.641 |
| Trust index | Religiosity | |||
|---|---|---|---|---|
| Low | High | Low | High | |
| (1) | (2) | (3) | (4) | |
| Corporate Integrity | −0.061** | −0.038 | −0.067** | −0.031 |
| (−2.059) | (−1.232) | (−2.227) | (−1.015) | |
| Beta Value-Weighted | 0.091* | 0.073 | 0.164*** | 0.011 |
| (1.710) | (1.497) | (2.959) | (0.238) | |
| Idiosyncratic Risk | 1.278*** | 0.304 | 1.014*** | 0.627*** |
| (5.628) | (1.439) | (4.504) | (2.878) | |
| Market Value of Equity | −0.434*** | −0.357*** | −0.318*** | −0.466*** |
| (−5.579) | (−5.163) | (−4.064) | (−6.861) | |
| Market-to-Book | −0.026*** | −0.034*** | −0.032*** | −0.030*** |
| (−3.066) | (−4.020) | (−3.882) | (−3.403) | |
| Book Leverage | 1.376*** | 2.013*** | 1.442*** | 1.941*** |
| (4.141) | (7.045) | (4.348) | (6.756) | |
| Momentum | −0.361*** | −0.258*** | −0.420*** | −0.215*** |
| (−5.595) | (−4.216) | (−6.400) | (−3.630) | |
| Analyst Forecast Dispersion | −0.111 | 0.008 | −0.162 | 0.033 |
| (−0.953) | (0.076) | (−1.303) | (0.341) | |
| Long-Term Growth Rate | 2.059*** | 1.376*** | 2.496*** | 0.948*** |
| (5.194) | (3.502) | (5.937) | (2.625) | |
| Return on Assets | −1.670*** | −1.497*** | −1.854*** | −1.362*** |
| (−3.126) | (−3.006) | (−3.181) | (−2.967) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 9,099 | 8,811 | 8,854 | 9,056 |
| Adjusted R2 | 0.614 | 0.646 | 0.618 | 0.641 |
Note(s): The table presents the relation between corporate integrity culture and the implied cost of equity capital. The dependent variable is the implied cost of equity capital. The independent variable of interest is Corporate Integrity. Other variables are defined in Appendix A. The regressions also include year fixed effects and firm fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
4.4 Endogeneity issues
4.4.1 Two-stage least squares
Endogeneity arising from unobserved omitted variables is possible. For instance, we are unable to assess whether the relation between corporate integrity culture and the cost of equity capital is confounded by the presence of an additional variable that has not been included in the analysis, such as management style. The management style of a company's leadership can influence both the cost of equity capital and corporate integrity culture. Therefore, our next step is to assess the robustness of our key findings by controlling for endogeneity through a 2SLS approach. This approach relies on the premise that our instrumental variables are associated with corporate integrity culture measures but are not associated with the error terms. We choose Distance to Religious Sites, the natural log of the geographic distance between a firm's headquarter and its closest religious site, as an instrument [22], [23] Using data from Tripadvisor, we identify religious sites based on three criteria: (1) The site must be a recognized religious landmark; (2) The religious site must have a rating of 4 or higher; (3) The religious site must have over 100 user reviews [24], [25] We then calculate the geographic distance between a firm's headquarter and these religious sites using their zip codes. The distance between latitude and longitude coordinates is determined using Vincenty's formula [26]. We argue that the distance between a firm and its closest religious site has a direct effect on the religiosity of a firm's headquarters, but it does not affect a firm's cost of equity.
Prior literature has shown that religiosity is a predictor of corporate integrity culture (McGuire et al., 2012) [27]. We argue that religiosity (and a firm's corporate integrity culture) will decrease when Distance to Religious Sites increases (the relevance condition) for a few reasons. First, the distance to highly rated religious sites is strongly correlated with local religiosity. Proximity to these sites enhances exposure and engagement with religious practices and traditions, making it more likely for individuals to become adherents or deepen their faith. As noted, religious sites act as hubs that strengthen knowledge of religious teachings, symbols and traditions among residents. Second, previous research (e.g. Williams, Francis, Robbins, & Annis, 2007; Sharpley & Sundaram, 2005) highlights that these sites influence a diverse audience, including local residents, tourists and non-adherents, encouraging broader participation in religious practices. In addition, Distance to Religious Sites does not have a direct effect on a firm's cost of equity capital (the exogeneity condition). In the first stage, we regress the corporate integrity culture on the instrumental variable, Distance to Religious Sites, together with a full set of controls as follows:
First stage:
where is Distance to Religious Sites. We include year-fixed effects and industry-fixed effects in the first stage [28]. Other control variables are the same as those in Eq. (1). In the second stage, we regress the cost of equity capital on the projected corporate integrity culture measure ), together with a full set of controls, as follows:
Second stage:
Table 6 Column (1) presents the results of the first-stage regression analysis in which we regress the corporate integrity culture measure on our instrumental variable. Consistent with our prediction, Distance to Religious Sites is negatively related to corporate integrity culture. The second-stage analysis in Column (2) reveals a negative and significant correlation between the predicted corporate integrity culture measure and a firm's cost of equity capital. The results reported in Table 6 are consistent with the view that corporate integrity culture is negatively associated with a firm's cost of equity capital after addressing potential endogeneity concerns through an instrumental-variable approach [29].
Corporate integrity culture and the implied cost of equity capital: An instrumental variable approach
| Corporate integrity | ICC | Corporate integrity dummy | ICC | |
|---|---|---|---|---|
| (1) | (2) | (3) | (4) | |
| Distance to religious sites | −0.052*** | −0.022*** | ||
| (−3.084) | (−2.921) | |||
| −0.026** | ||||
| (−2.565) | ||||
| −0.156** | ||||
| (−2.565) | ||||
| Beta value-weighted | −0.105*** | 0.191* | −0.043*** | 0.583** |
| (−7.743) | (1.705) | (−6.795) | (2.228) | |
| Idiosyncratic risk | 0.507*** | 2.107*** | 0.127*** | 2.779*** |
| (8.960) | (3.926) | (5.315) | (3.509) | |
| Market value of equity | 0.075*** | −0.203** | 0.018*** | −0.125 |
| (12.001) | (−1.987) | (6.819) | (−0.949) | |
| Market-to-book | 0.005*** | −0.019*** | 0.002** | −0.002 |
| (2.947) | (−2.876) | (2.485) | (−0.172) | |
| Book leverage | −0.134*** | 1.362*** | −0.067*** | 0.671 |
| (−3.091) | (6.702) | (−3.339) | (1.626) | |
| Momentum | −0.096*** | −0.555*** | −0.032*** | −0.811*** |
| (−5.300) | (−5.329) | (−3.919) | (−4.060) | |
| Analyst forecast dispersion | −0.001 | −0.053 | 0.013 | 0.152 |
| (−0.044) | (−0.713) | (1.116) | (1.382) | |
| Long-term growth rate | −0.037 | 1.641*** | 0.025 | 2.122*** |
| (−0.495) | (6.305) | (0.687) | (7.972) | |
| Return on assets | −0.414*** | −2.633*** | −0.170*** | −4.218*** |
| (−3.941) | (−4.804) | (−3.464) | (−3.809) | |
| Year fixed effects | Yes | Yes | Yes | Yes |
| Industry fixed effects | Yes | No | Yes | No |
| Firm fixed effects | No | Yes | No | Yes |
| Observations | 17,910 | 17,910 | 17,910 | 17,910 |
| Adjusted R2 | 0.869 | 0.629 | 0.275 | 0.629 |
| Corporate integrity | ICC | Corporate integrity dummy | ICC | |
|---|---|---|---|---|
| (1) | (2) | (3) | (4) | |
| Distance to religious sites | −0.052*** | −0.022*** | ||
| (−3.084) | (−2.921) | |||
| −0.026** | ||||
| (−2.565) | ||||
| −0.156** | ||||
| (−2.565) | ||||
| Beta value-weighted | −0.105*** | 0.191* | −0.043*** | 0.583** |
| (−7.743) | (1.705) | (−6.795) | (2.228) | |
| Idiosyncratic risk | 0.507*** | 2.107*** | 0.127*** | 2.779*** |
| (8.960) | (3.926) | (5.315) | (3.509) | |
| Market value of equity | 0.075*** | −0.203** | 0.018*** | −0.125 |
| (12.001) | (−1.987) | (6.819) | (−0.949) | |
| Market-to-book | 0.005*** | −0.019*** | 0.002** | −0.002 |
| (2.947) | (−2.876) | (2.485) | (−0.172) | |
| Book leverage | −0.134*** | 1.362*** | −0.067*** | 0.671 |
| (−3.091) | (6.702) | (−3.339) | (1.626) | |
| Momentum | −0.096*** | −0.555*** | −0.032*** | −0.811*** |
| (−5.300) | (−5.329) | (−3.919) | (−4.060) | |
| Analyst forecast dispersion | −0.001 | −0.053 | 0.013 | 0.152 |
| (−0.044) | (−0.713) | (1.116) | (1.382) | |
| Long-term growth rate | −0.037 | 1.641*** | 0.025 | 2.122*** |
| (−0.495) | (6.305) | (0.687) | (7.972) | |
| Return on assets | −0.414*** | −2.633*** | −0.170*** | −4.218*** |
| (−3.941) | (−4.804) | (−3.464) | (−3.809) | |
| Year fixed effects | Yes | Yes | Yes | Yes |
| Industry fixed effects | Yes | No | Yes | No |
| Firm fixed effects | No | Yes | No | Yes |
| Observations | 17,910 | 17,910 | 17,910 | 17,910 |
| Adjusted R2 | 0.869 | 0.629 | 0.275 | 0.629 |
Note(s): The table presents the 2SLS regression results. The instrumental variable, Distance to Religious Sites, is the distance between a firm and closest religious site. The dependent variables in the first stage are Corporate Integrity in Column (1) and Corporate Integrity Dummy in Column (3). The first-stage regressions include year fixed effects and industry fixed effects. The dependent variable in the second stage is the implied cost of equity capital. The key independent variables, Corporate Integrity and Corporate Integrity Dummy in Columns (2) and (4) are predicted by the first stage. The second-stage regressions include year fixed effects and firm fixed effects. Other variables are defined in Appendix A. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
4.4.2 Propensity score matching
To further control for any endogenous selection, we use a PSM approach to construct a control sample based on all the control variables. We match without replacement and require the propensity scores for each matched pair to be within 0.01 of each other. In Table 7, we match 3,388 firms with strong integrity culture scores with 3,388 firms with weak integrity culture scores, resulting in a sample of 6,776 firm-year observations.
Corporate integrity culture and the implied cost of equity capital: Propensity score matching
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity | −0.080** | −0.082** | ||
| (−2.366) | (−2.425) | |||
| Corporate Integrity Dummy | −0.320*** | −0.323*** | ||
| (−4.678) | (−4.733) | |||
| Beta Value-Weighted | −0.083 | −0.080 | ||
| (−1.168) | (−1.138) | |||
| Idiosyncratic Risk | 0.903*** | 0.914*** | ||
| (2.790) | (2.827) | |||
| Market Value of Equity | −0.413*** | −0.356*** | −0.414*** | −0.357*** |
| (−4.082) | (−3.477) | (−4.108) | (−3.494) | |
| Market-to-Book | −0.020* | −0.021* | −0.020* | −0.021* |
| (−1.775) | (−1.891) | (−1.815) | (−1.934) | |
| Book Leverage | 1.779*** | 1.755*** | 1.780*** | 1.755*** |
| (4.279) | (4.247) | (4.291) | (4.257) | |
| Momentum | −0.399*** | −0.436*** | −0.394*** | −0.432*** |
| (−4.380) | (−4.766) | (−4.341) | (−4.735) | |
| Analyst Forecast Dispersion | 0.043 | 0.029 | 0.035 | 0.021 |
| (0.283) | (0.191) | (0.232) | (0.138) | |
| Long-Term Growth Rate | 1.133* | 1.110* | 1.118* | 1.094* |
| (1.913) | (1.875) | (1.892) | (1.853) | |
| Return on Assets | −1.641** | −1.508** | −1.652** | −1.517** |
| (−2.263) | (−2.074) | (−2.284) | (−2.091) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 6,776 | 6,776 | 6,776 | 6,776 |
| Adjusted R2 | 0.642 | 0.643 | 0.644 | 0.645 |
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity | −0.080** | −0.082** | ||
| (−2.366) | (−2.425) | |||
| Corporate Integrity Dummy | −0.320*** | −0.323*** | ||
| (−4.678) | (−4.733) | |||
| Beta Value-Weighted | −0.083 | −0.080 | ||
| (−1.168) | (−1.138) | |||
| Idiosyncratic Risk | 0.903*** | 0.914*** | ||
| (2.790) | (2.827) | |||
| Market Value of Equity | −0.413*** | −0.356*** | −0.414*** | −0.357*** |
| (−4.082) | (−3.477) | (−4.108) | (−3.494) | |
| Market-to-Book | −0.020* | −0.021* | −0.020* | −0.021* |
| (−1.775) | (−1.891) | (−1.815) | (−1.934) | |
| Book Leverage | 1.779*** | 1.755*** | 1.780*** | 1.755*** |
| (4.279) | (4.247) | (4.291) | (4.257) | |
| Momentum | −0.399*** | −0.436*** | −0.394*** | −0.432*** |
| (−4.380) | (−4.766) | (−4.341) | (−4.735) | |
| Analyst Forecast Dispersion | 0.043 | 0.029 | 0.035 | 0.021 |
| (0.283) | (0.191) | (0.232) | (0.138) | |
| Long-Term Growth Rate | 1.133* | 1.110* | 1.118* | 1.094* |
| (1.913) | (1.875) | (1.892) | (1.853) | |
| Return on Assets | −1.641** | −1.508** | −1.652** | −1.517** |
| (−2.263) | (−2.074) | (−2.284) | (−2.091) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 6,776 | 6,776 | 6,776 | 6,776 |
| Adjusted R2 | 0.642 | 0.643 | 0.644 | 0.645 |
Note(s): This table presents the relation between corporate integrity culture and the implied cost of equity capital using propensity score matching. The dependent variable is the implied cost of equity capital. The independent variables of interest are Corporate Integrity in Columns (1) and (2) Corporate Integrity Dummy in Columns (3) and (4). Other variables are defined in Appendix A. The regressions also include year fixed effects and firm fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
The insights derived from Table 7 align closely with those from Table 2. To ensure that firms with strong corporate integrity and those with weak corporate integrity share similar characteristics, we perform a mean comparison test between PSM strong corporate integrity firms and PSM control firms. Appendix F Panel A demonstrates that, prior to PSM matching, firm characteristics differ significantly between firms with a strong integrity culture and those with a weak integrity culture in the full sample. In contrast, Appendix F Panel B reveals that, after PSM matching, the differences in firm characteristics between these two groups become insignificant in the reduced sample.
4.4.3 Reverse causality issue
Up to this point, our results show that corporate integrity culture is negatively associated with the cost of equity capital. One could argue that firms facing a high cost of equity due to external factors might begin “cutting corners”, thereby undermining a culture of integrity. To counter this argument, we examine the lead-lag relation and demonstrate that changes in culture precede changes in the cost of equity capital. Specifically, we use the lagged corporate integrity culture as the independent variable to further examine whether corporate integrity culture in the previous year affects the cost of equity capital in the current year. All control variables are also lagged in our regressions. Table 8 presents the relation between the corporate integrity culture at year t-1 and the implied cost of equity at year t. Similar to the results shown in Table 2, we find that corporate integrity culture at year t-1 is negatively related to the implied cost of equity at year t.
Lagged corporate integrity culture and the implied cost of equity capital
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity t-1 | −0.023* | −0.026* | ||
| (−1.681) | (−1.686) | |||
| Corporate Integrity Dummy t-1 | −0.078* | −0.078* | ||
| (−1.713) | (−1.709) | |||
| Beta Value-Weighted t-1 | 0.056 | 0.056 | ||
| (1.288) | (1.282) | |||
| Idiosyncratic Risk t-1 | 0.575*** | 0.571*** | ||
| (3.176) | (3.163) | |||
| Market Value of Equity t-1 | −0.360*** | −0.326*** | −0.361*** | −0.327*** |
| (−5.885) | (−5.224) | (−5.906) | (−5.244) | |
| Market-to-Book t-1 | −0.029*** | −0.030*** | −0.029*** | −0.030*** |
| (−3.882) | (−4.015) | (−3.885) | (−4.017) | |
| Book Leverage t-1 | 1.540*** | 1.517*** | 1.541*** | 1.518*** |
| (6.351) | (6.263) | (6.355) | (6.265) | |
| Momentum t-1 | −0.247*** | −0.274*** | −0.247*** | −0.274*** |
| (−4.843) | (−5.331) | (−4.854) | (−5.338) | |
| Analyst Forecast Dispersion t-1 | −0.035 | −0.058 | −0.033 | −0.056 |
| (−0.363) | (−0.597) | (−0.345) | (−0.577) | |
| Long-Term Growth Rate t-1 | 1.876*** | 1.826*** | 1.875*** | 1.826*** |
| (5.522) | (5.386) | (5.528) | (5.395) | |
| Return on Assets t-1 | −1.295*** | −1.159*** | −1.296*** | −1.161*** |
| (−3.085) | (−2.762) | (−3.087) | (−2.764) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 14,489 | 14,489 | 14,489 | 14,489 |
| Adjusted R2 | 0.646 | 0.646 | 0.646 | 0.646 |
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| Corporate Integrity t-1 | −0.023* | −0.026* | ||
| (−1.681) | (−1.686) | |||
| Corporate Integrity Dummy t-1 | −0.078* | −0.078* | ||
| (−1.713) | (−1.709) | |||
| Beta Value-Weighted t-1 | 0.056 | 0.056 | ||
| (1.288) | (1.282) | |||
| Idiosyncratic Risk t-1 | 0.575*** | 0.571*** | ||
| (3.176) | (3.163) | |||
| Market Value of Equity t-1 | −0.360*** | −0.326*** | −0.361*** | −0.327*** |
| (−5.885) | (−5.224) | (−5.906) | (−5.244) | |
| Market-to-Book t-1 | −0.029*** | −0.030*** | −0.029*** | −0.030*** |
| (−3.882) | (−4.015) | (−3.885) | (−4.017) | |
| Book Leverage t-1 | 1.540*** | 1.517*** | 1.541*** | 1.518*** |
| (6.351) | (6.263) | (6.355) | (6.265) | |
| Momentum t-1 | −0.247*** | −0.274*** | −0.247*** | −0.274*** |
| (−4.843) | (−5.331) | (−4.854) | (−5.338) | |
| Analyst Forecast Dispersion t-1 | −0.035 | −0.058 | −0.033 | −0.056 |
| (−0.363) | (−0.597) | (−0.345) | (−0.577) | |
| Long-Term Growth Rate t-1 | 1.876*** | 1.826*** | 1.875*** | 1.826*** |
| (5.522) | (5.386) | (5.528) | (5.395) | |
| Return on Assets t-1 | −1.295*** | −1.159*** | −1.296*** | −1.161*** |
| (−3.085) | (−2.762) | (−3.087) | (−2.764) | |
| Year Fixed Effects | Yes | Yes | Yes | Yes |
| Firm Fixed Effects | Yes | Yes | Yes | Yes |
| Observations | 14,489 | 14,489 | 14,489 | 14,489 |
| Adjusted R2 | 0.646 | 0.646 | 0.646 | 0.646 |
Note(s): This table presents the relation between corporate integrity culture and the implied cost of equity capital. The dependent variable is the implied cost of equity capital. The independent variables of interest are Corporate Integrity in Columns (1) and (2) and Corporate Integrity Dummy in Columns (3) and (4). Other variables are defined in Appendix A. The corporate integrity measure and other control variables are from the past year. The regressions also include year fixed effects and firm fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
4.5 Alternative measures of the cost of equity capital
In our baseline analysis, we employ four estimates of the implied cost of equity capital, which deduce the ex ante cost of equity capital that is implied by share prices and analyst earnings forecasts. Thus, unlike traditional methods of assessing the cost of equity capital (such as single-factor or multi-factor asset pricing models), these four models do not rely on historical data and avoid the use of unstable realized returns to derive precise estimates of the firm-level cost of equity capital (Pástor, Sinha, & Swaminathan, 2008). However, existing evidence suggests that accounting-based measures of the ICC may not be reliable under certain scenarios (Easton & Monahan, 2005). Consistent with this view, Lee, So, and Wang (2021) demonstrate that accounting-based ICCs underperform characteristic-based measures of the cost of equity capital in the cross-section. Previous literature also uses risk factor-based cost of equity capital estimates as a proxy of the cost of equity capital (Ashbaugh-Skaife et al., 2009; Barth, Konchitchki, & Landsman, 2013; Lara, García Osma, & Penalva, 2011; Li & Mohanram, 2014; McInnis, 2010; Mohanram & Rajgopal, 2009). Therefore, to alleviate these concerns, we rerun the baseline regression using the expected cost of equity capital estimated based on the Fama-French and momentum factors [30]. Table 9 provides robust evidence that corporate integrity is significantly and negatively associated with the expected cost of equity capital. The inclusion of various control variables across multiple models ensures that these findings are reliable and consistent.
Corporate integrity culture and expected cost of equity capital
| (1) | (2) | |
|---|---|---|
| Corporate integrity | −0.441** | |
| (−2.531) | ||
| Corporate integrity dummy | −1.022*** | |
| (−2.985) | ||
| Beta value-weighted | 2.917*** | 2.915*** |
| (9.948) | (9.959) | |
| Idiosyncratic risk | 17.215*** | 17.115*** |
| (12.575) | (12.610) | |
| Market value of equity | −0.248 | −0.265 |
| (−1.414) | (−1.527) | |
| Market-to-book | 0.360*** | 0.360*** |
| (8.903) | (8.917) | |
| Book leverage | −3.603*** | −3.609*** |
| (−2.927) | (−2.935) | |
| Momentum | 5.929*** | 5.936*** |
| (18.663) | (18.684) | |
| Analyst forecast dispersion | −5.050*** | −5.042*** |
| (−8.904) | (−8.890) | |
| Long-term growth rate | 35.644*** | 35.690*** |
| (18.097) | (18.147) | |
| Return on assets | 50.599*** | 50.610*** |
| (20.811) | (20.818) | |
| Year fixed effects | Yes | Yes |
| Industry fixed effects | Yes | Yes |
| Observations | 17,910 | 17,910 |
| Adjusted R2 | 0.765 | 0.765 |
| (1) | (2) | |
|---|---|---|
| Corporate integrity | −0.441** | |
| (−2.531) | ||
| Corporate integrity dummy | −1.022*** | |
| (−2.985) | ||
| Beta value-weighted | 2.917*** | 2.915*** |
| (9.948) | (9.959) | |
| Idiosyncratic risk | 17.215*** | 17.115*** |
| (12.575) | (12.610) | |
| Market value of equity | −0.248 | −0.265 |
| (−1.414) | (−1.527) | |
| Market-to-book | 0.360*** | 0.360*** |
| (8.903) | (8.917) | |
| Book leverage | −3.603*** | −3.609*** |
| (−2.927) | (−2.935) | |
| Momentum | 5.929*** | 5.936*** |
| (18.663) | (18.684) | |
| Analyst forecast dispersion | −5.050*** | −5.042*** |
| (−8.904) | (−8.890) | |
| Long-term growth rate | 35.644*** | 35.690*** |
| (18.097) | (18.147) | |
| Return on assets | 50.599*** | 50.610*** |
| (20.811) | (20.818) | |
| Year fixed effects | Yes | Yes |
| Industry fixed effects | Yes | Yes |
| Observations | 17,910 | 17,910 |
| Adjusted R2 | 0.765 | 0.765 |
Note(s): The table presents the relation between corporate integrity culture and the expected cost of equity capital. The dependent variable is the expected cost of equity capital estimated using the Fama-French and momentum factors. The independent variables of interest are Corporate Integrity in Column (1) and Corporate Integrity Dummy in Column (2). Other variables are defined in Appendix A. The regressions also include year fixed effects and industry fixed effects. t-statistics is based on standard errors adjusted for heteroscedasticity and clustered at the firm level. *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively
5. Conclusions
This study provides robust evidence that corporate integrity culture plays a vital role in reducing the cost of equity capital. Using firm-level integrity data, we demonstrate that a strong culture of integrity reduces the risk of accounting malfeasance, which serves as a critical channel for lowering the cost of equity capital, thereby underscoring integrity's role in promoting transparency and trustworthiness within organizations. Our cross-sectional analysis further highlights that the benefits of integrity culture are particularly pronounced in low-trust and low-religiosity environments, where the capital market rewards firms for fostering trust.
This study makes several important contributions to the literature. First, it is the first to empirically investigate whether corporate integrity culture enhances firm value by reducing the cost of equity capital and establishing a previously unexplored mechanism. Second, this study contributes to the existing literature on factors affecting the cost of equity capital by identifying integrity culture as a key factor and emphasizing the role of accounting malfeasance risk as a channel through which it influences the cost of equity capital. Third, using firm-level integrity culture data, this study addresses gaps in the finance and accounting literature, where integrity culture has received limited attention despite its significance. Using firm-level data enables a more precise and nuanced analysis, overcoming limitations of traditional proxies such as surveys or regional religious measures. This research broadens the understanding of how corporate integrity culture shapes financial outcomes and provides valuable insights for both academic and practical applications.
Notes
An independent senior advisor to Deloitte and Touche said, “In the long run, a positive culture of integrity is the foundation for an effective ethics and compliance program, which, when properly embedded into an organization, can create a competitive advantage and serve as a valuable organizational asset,“; and “Building a culture of integrity not only fortifies the organization against risk, but also builds both employee engagement and strong loyalties with all stakeholders.” Therefore, the business world understands the important role of integrity.
We focus on the cost of equity because the Gordon Growth Model (P = D1/r-g) suggests that a firm's current stock price (P) is a function of the expected dividend next year (D1), cost of equity capital (r) and constant dividend growth rate (g). In this paper, we use the cost of equity capital, the cost of equity capital financing and the implied cost of equity capital interchangeably.
Upon further analysis (untabulated), we identify that the negative effect also holds for the composite measure of culture and three other cultural dimensions: innovation, respect and quality. This is not surprising given Li et al. (2021)'s finding that these cultural dimensions are correlated.
In a survey of 1,348 North American executives, 92% expressed the belief that enhancing the corporate culture would enhance the value of their firm (Graham et al., 2022). Their study builds upon the findings of Guiso et al. (2015) by exploring both the corporate websites and the employees' perspectives on company culture, including both the ideal values and daily practices. Furthermore, their examination of the executives' views aligns with the relationship between integrity and firm value documented in Guiso et al. (2015). However, they also discover that corporate websites provide limited information on the firm's value.
In Appendix A, we list the 30 most representative words (seed words) for integrity culture which the word-embedding model uses to derive the measure of corporate culture (Li et al., 2021). Please refer to Li et al. (2021) for the details of the model.
We argue that culture is not the same as other formal monitoring mechanisms such as corporate governance and internal controls. Integrity culture reflects an informal system of shared ethical norms that shapes behavior throughout the organization. Because informal norms can influence managerial behavior even when formal monitoring is incomplete, integrity culture may have incremental implications for information risk and financing costs beyond those captured by traditional governance mechanisms. Although integrity culture and CSR both reflect aspects of corporate values, they represent conceptually distinct constructs. CSR primarily captures a firm's commitment to fulfilling its responsibilities toward external stakeholders through social, environmental and community-oriented activities. In contrast, integrity culture reflects internal organizational norms emphasizing honesty, truthfulness and ethical conduct in day-to-day decision-making.
For instance, EY Resigns as Auditor for Super Micro, Citing Concerns Over ‘Integrity’ (Link to the website).
Corporate integrity data can be obtained from Link to the website.
We limit our sample to US firms because the data we use for the 2SLS approach and cross-sectional analysis are only available for US firms.
Appendix D displays the correlations between the estimates of the implied cost of equity, with Pearson correlations presented below the diagonal and Spearman correlations presented above the diagonal.
We use the implied cost of equity capital estimates rather than other proxies for the cost of equity capital for three reasons: (1) Elton (1999) argues that implied cost of equity estimates align with the principles of expected return theory, as they are forward-looking, while average realized returns do not fit this theory well. (2) Hail & Leuz (2006) note that implied cost-of-equity estimates avoid the problem of selecting a market portfolio that arises in stock return-based studies. (3) Pástor et al. (2008) find that the implied cost-of-equity capital captures almost 90% of the variation in the true cost of capital, suggesting that these methods are suitable for research purposes when the cost of capital is required as a variable in the cross-sectional analysis.
Li et al. (2021) explain that they focus on the Q&A section rather than the scripted management presentation to measure corporate culture, as this helps mitigate excessive self-promotion; managers have limited control over which topics are discussed during the Q&A. However, a potential drawback of this approach is that managers may only address aspects of their culture if prompted by a participant's question. Consequently, insights into a firm's integrity culture may be limited unless a relevant question arises.
In our study, to make the scale of the results more suitable for analysis and interpretation, we multiplied the ICC by 100, following El Ghoul et al. (2011).
Please refer to Appendix B for variable definitions.
The market beta is computed by conducting a regression analysis that involves regressing the daily stock returns of individual stocks over a fiscal year against the concurrent CRSP value-weighted market returns. This analysis considers non-synchronous trading, as per the method introduced by Scholes and Williams (1977). In a similar vein, we define idiosyncratic risk as the yearly standard deviation of the residuals derived from the above-mentioned regression analysis that is adjusted for non-synchronous trading, utilizing the methodology put forth by Scholes and Williams (1977).
For a one-standard-deviation change in Corporate Integrity, calculate the economic impact by multiplying the coefficient by the standard deviation of Corporate Integrity from Table 1 Panel B (−0.050 × 0.888 = −0.0444). The economic significance of corporate integrity is material. A one-standard-deviation improvement in Corporate Integrity results in a modest but meaningful reduction in equity costs.
Firms classified as having strong corporate integrity (compared to weak) have an implied cost of equity capital that is approximately 10.6–11.0 basis points lower (Table 2 Columns (3) and (4)).
See Li et al. (2021) Table 5 Panel B on page 3289.
We do not use religiosity as the instrumental variable because it fails the exogeneity condition. For instance, Hilary and Hui (2009) find that the level of religiosity of a firm's headquarters affects the focal firm's stock return. Hu, Chen, and Liu (2018) find that religious atmosphere affects the cost of equity capital.
We intentionally do not use direct measures of religiosity as an instrument because prior literature suggests that religiosity may directly influence financial reporting outcomes and financing costs (e.g. Hilary & Hui, 2009; Hu et al., 2018), potentially violating the exclusion restriction. Instead, we employ a more geographically remote determinant of the local religious environment that is less likely to directly affect a firm's cost of equity capital.
We collect religious sites data from Tripadvisor: Link to the website.
In Appendix E, we present the geographic distribution of religious sites in the United States.
Vincenty's formula, which calculates the geodesic distance (shortest distance on the ellipsoidal surface) between two points specified by latitude and longitude, is as follows:
where () and () are the latitude and longitude coordinates of the small business and the nearest highly rated religious site. Distance to Religious Sites is the natural log of Distance.
The religiosity level of where a firm's headquarters is located may have a positive effect on corporate integrity culture at the firm level because a strong religious social norm may have a spillover effect on a firm's managers and employees. Similarly, the studies below also show how religiosity can affect human behaviors.
We control for industry-fixed effects instead of firm-fixed effects to address multicollinearity concerns.
We acknowledge, however, that the validity of any instrumental-variable design ultimately depends on the exclusion restriction, which cannot be tested directly and must rely on theoretical justification (Larcker & Rusticus, 2010). Although no instrument can be proven to be perfectly exogenous, we believe that distance to religious sites represents a reasonable and informative instrument because it is conceptually distinct from direct measures of religiosity that have been shown to affect financing outcomes.
We show the calculation of the expected cost of equity capital in Appendix G.
The supplementary material for this article can be found online.

