Religiosity at the Top and Annual Report Readability
Accounting and Information Systems Department, The University of Texas at El Paso, El Paso, TX 79968, USA
*
Author to whom correspondence should be addressed.
J. Risk Financial Manag. 2022, 15(10), 485; https://doi.org/10.3390/jrfm15100485
Submission received: 19 September 2022
/
Revised: 15 October 2022
/
Accepted: 18 October 2022
/
Published: 21 October 2022
Abstract
:This paper examines how individual religiosity at the top level of organizations affects the quality of their disclosure practices, as measured by the readability of annual reports. Our paper extends the recent accounting and finance literature that moves away from a location-based measure to an individual-based measure for capturing the effect of religiosity. Our findings suggest that the individual religiosity of C-suite executives matters in corporate decision-making and has positive implications for the quality of corporate disclosure practices, as reflected by more readable reports. This main finding is primarily driven by the religiosity of CEOs. Additional findings also suggest that the effect of religiosity is not solely driven by the religious denomination of the majority group within a given location-based setting. Previous research using religiosity proxies based on the majority religion in the locale of firms’ headquarters may have measurement issues that disguise the effect of religiosity. This issue is particularly problematic when CEOs or other executives participate in minority religious denominations. Overall, our paper finds that CEO religiosity is an important attribute that affects the overall quality of business practice.
1. Introduction & Literature Review
How important is top-management religiosity, and how does it impact business decision-making and information-sharing transparency? The topic of religion has been studied in varying capacities within multiple sub-disciplines of the broader social science literature since the late 1700s (Smith 1776). One potential explanation for this multi-stream focus lies in the connection between religion and human rationale (Smith 1776; Jung 1960). This connection has significant economic, psychological, and social implications for modern-day norms, behavior, and business practices (Anderson 1988; Wulff 1991; Sunstein 1996; Beit-Hallahmi and Argyle 1997; Iannaccone 1998; Kennedy and Lawton 1998; Guiso et al. 2003; Stulz and Williamson 2003). These far-reaching implications of religion on society and business may potentially explain the multiple sub-disciplinary foci of religion within the social sciences.
Recent studies examining religion’s influence on business practices are generally clustered within the accounting or finance domain. These studies typically use location-based proxies for religiosity based on firm headquarters locations to examine how religion affects the quality of accounting, financial reporting practices, financial decision-making, or other observable indicators of quality of business practices. The consensus within the literature appears to be that religion has positive firm-level implications. In other words, firms that have headquarters in highly religious locations have less litigation, lower restatement rates, higher quality of management forecasts, higher quality of accruals, and less opportunistic earnings management (Grullon et al. 2009; Dyreng et al. 2012; McGuire et al. 2012; Du et al. 2015; Chourou et al. 2020). Measures of firm headquarter location-based religiosity also have been associated with positive audit-related implications for firms, as evidenced by the lower number of going-concern reports issued by auditors and lower audit fees charged by auditors to clients who are based in highly religious areas (Leventis et al. 2018; Omer et al. 2018). Corporations based in religious areas generally find it easier to raise debt financing (Chen et al. 2016a; He and Hu 2016; Jiang et al. 2018; Cai and Shi 2019). Corporations headquartered in religious areas also tend to make less risky decisions as evidenced by their greater focus on long-term growth, litigation risk minimization, shareholder wealth creation, and less risky innovation (Hilary and Hui 2009; Al Rahahleh et al. 2019; Ma et al. 2020).
However, this location-based approach to measuring religiosity has limitations, which has created another avenue for research in recent years. Location-based studies assume that the religious beliefs and viewpoints of the majority of the inhabitants in a county are reflected in corporate decisions. However, these extant studies fail to demonstrate how the locational effect of religion translates to the firm level (Baxamusa and Jalal 2016; Cai et al. 2019; Chen et al. 2022). In other words, these research papers fail to consider the religious affiliations of the firm’s decision-makers themselves (Baxamusa and Jalal 2016; Cai et al. 2019; Chen et al. 2022). Thus, the potential for misclassification of the religiosity of firm decision makers exists if a firm is classified as being religious solely based on its geographic location when its CEO or other C-suite members are not religious. Location-based studies also ignore that a C-suite executive’s religion may differ from the affiliation of the majority of inhabitants of a given county (Baxamusa and Jalal 2016). For example, a positive firm-level implication may be attributed to Christianity, the dominant religion in most counties within the U.S.A. (America’s Changing Religious Landscape 2015). However, in reality, it could be the decision-making quality of a religious CEO belonging to a non-Christian religious denomination that, in essence, drives previous findings. Thus, capturing religiosity at the executive or decision-maker level helps to solve these misclassification issues associated with the location-based religiosity measure, as suggested by Baxamusa and Jalal (2016), Cai et al. (2019) and Chen et al. (2022). Capturing religiosity at the individual level also shows that the effect of religiosity is not solely driven by the adherents of the religious group that constitutes the majority within a given locale. Overall, this measurement issue involving religiosity constitutes one of the three core motivations for this study.
The second motivation stems from the connection between the broader disclosure literature and religion. Although recent studies have examined how religion affects disclosure practices, there are gaps in the literature. These studies generally examine how religion affects disclosure practices using disclosure length of annual report sections (Aribi and Gao 2012; Elamer et al. 2020), earnings management and timely loss recognition, or optimism of management earnings forecasts (Chourou et al. 2020; Oh and Shin 2020). One concurrent working paper examines the relationship between religion and disclosure readability (Cano-Rodríguez and Moreno 2020), but it measures religiosity using a less-accurate, location-based proxy. Hence, to the best of our knowledge, our research is the first to employ an individual-based measure of top management religiosity to study how the religiosity of firm decision-makers affects the readability of firm disclosures.
Within the realms of the individual religiosity literature in accounting and finance, our paper is one of the first studies to utilize publicly available data from BoardEx to identify the religiosity and the religious denominations of executives through their membership in religious organizations. This approach of identifying both the religiosity and the religious denominations of executives is likely to be more comprehensive and direct compared to Cai et al. (2019) and Chen et al. (2022). These two studies identify executive faith and religiosity through the religious affiliations of an executive’s undergraduate or graduate degree issuing university. Such an approach could cause a mismatch where the religious denomination and/or religiosity of an executive does not match the religious denomination and/or religiosity of their undergraduate or graduate university. A religious executive who publicly discloses his/her religiosity, may not have attended a religious university, and therefore may not be accurately classified as per Cai et al. (2019) and Chen et al. (2022)’s approach. Additionally, our study expands the range of publicly available datasets on individual religiosity that can be utilized by future researchers. It provides a credible alternative to the “Marquis Who’s Who” magazine data that Baxamusa and Jalal (2016) utilize for gathering their data on individual religiosity. Overall, introducing and utilizing the benefits of a unique subsection of BoardEx data that contains publicly available information on the individual religiosity of C-suite executives, form another core motivation for this study.
Our study’s findings suggest that the individual religiosity of the firm’s top-level managers has important and positive implications for a firm’s disclosure quality, as measured by the readability of those firms’ annual reports. In subsample analyses, CEOs appear to be the drivers of this positive relationship between religiosity and readability compared to the other non-CEO executives. This finding is consistent with CEOs being at the center of firm-communication channels such as disclosure and annual report preparation-related decisions (Sherin 2010; Lee et al. 2012; Ke et al. 2019). Our main findings hold through multiple robustness tests using alternate samples matched on propensity scores and entropy-balancing firm and annual report characteristic-related controls. Additional subsample analysis also offers mixed evidence that executives with disclosed/identifiable religiosity from both Christian and non-Christian denominations significantly influence the readability levels of annual reports and the overall disclosure practices of firms. This finding is another novel contribution of our study as prior literature argues that the religion of the majority within a county in which a firm is located drives any positive associations between the overall religiosity of a county and the quality of accounting, financial reporting, or other business practices (Grullon et al. 2009; Dyreng et al. 2012; McGuire et al. 2012; Du et al. 2015; Leventis et al. 2018; Omer et al. 2018). Overall, our results help extend the scope of both the broader religiosity literature and the broader disclosure literature involving readability.
2. Hypothesis
The existing psychology and theological literature argue that two distinct viewpoints explain how humans use their religious identities when making decisions—the intrinsic dimension point of view and the extrinsic dimension perspective (Salsman et al. 2005; Vitell et al. 2009). Arguments from both these viewpoints affect how humans communicate information to one another. Firm financial disclosure is a common way corporate executives communicate with firm stakeholders. Thus, these different viewpoints provide a framework to predict how the religiosity of these executives affects corporate communication through disclosure.
The intrinsic dimension viewpoint of religiosity argues that humans rely on time-invariant forms of ascetic morality and do not increase or reduce their religiousness based on extrinsic needs (Middleton and Putney 1962; Salsman et al. 2005; Hardy 2006; Vitell et al. 2009). Furthermore, this perspective argues that religious individuals are highly morally conscious of the value implications of their actions to the overall society (Geyer and Baumeister 2005; Salsman et al. 2005; Vitell et al. 2009). Hence, based on the intrinsic dimension viewpoint, religious individuals try to avoid as much uncertainty and be as transparent as possible in their actions, as such choices maximize the overall value-related implications of their actions for the entire community to which they belong (Roccas 2005). This high moral consciousness and desire to be transparent for the common good is primarily driven by the sacredness with which religious individuals value the core of their belief systems (Johnson 1959). Philosophically, this would be similar to honorable merchants in ancient societies, who were viewed as highly rational individuals who were dedicated and committed to their responsibilities and duty of care in long-term business relationships (Milkau 2017; Bott and Milkau 2018). Thus, proponents of this intrinsic dimension-oriented viewpoint of theology and psychology argue that religious individuals generally do not try to mislead their community.
The intrinsic dimension perspective suggests that more religious individuals in executive positions will create financial disclosures that are not purposefully misleading to financial statement users. Annual reports contain mandatory and voluntary disclosures that help explain a firm’s performance and business practices beyond what is provided in summary financial statements (Loughran and McDonald 2011, 2014, 2016). Purposefully distorting this disclosed information has negative value implications for a firm’s stakeholders (Gibbins et al. 1990). The opportunistic manipulation of information in annual reports would be similar to a purposeful distortion of the character and sanctity of religious fundamentals and core belief systems under the intrinsic dimension viewpoint (Johnson 1959; Pargament et al. 2005). Hence, given the arguments set forth by the proponents of the intrinsic dimension viewpoint, the religiosity of a firm’s C-suite executives is likely to be associated with increased annual report readability.1
Conversely, the extrinsic dimension viewpoint of psychology and theology suggests that increased religiosity may be used to reduce transparency in financial reporting. This perspective advocates that humans increase or reduce their religiousness based on extrinsic needs (Salsman et al. 2005; Vitell et al. 2009). In other words, individuals use their religious identity opportunistically depending on the situation or context they are in. Differences in religious interpretations that exist between sects or religious scholars within a particular religious subgroup (Adhikari and Agrawal 2016; Baxamusa and Jalal 2016; Bhatti 2019), may amplify this religious identity-related opportunism and encourage individuals to undertake more risks (Salsman et al. 2005; Vitell et al. 2009). Rawwas et al. (2006) and Li (2008) provide evidence of religious identity-induced opportunism within the business literature by documenting that religious identity may be strategically marketed to fit one’s personal or business-related objectives and needs. Hence, managers may use their religiosity as a form of window-dressing or cheap talk (Lyons and Mehta 1997; Chen et al. 2016b; Lizińska and Czapiewski 2019) to hide subpar business performance. These actions are similar to how superstar CEOs with media limelight use their media coverage as an attention-diverting tool (Malmendier and Tate 2009). In other words, executives may masquerade their religiosity to cover for their deliberate and opportunistic opacity in firm disclosures (Rawwas et al. 2006; Li 2008).
Overall, the combined implications from both the intrinsic and extrinsic dimension perspectives suggest that religiosity at the top level of an organization could influence the degree of transparency of firm-level communications, such as in annual reports. Although the direction of this influence could be positive or negative, it can have important implications for observable measures of transparency such as readability. These arguments lead to our hypothesis.
H1.
Religiosity at the top influences a firm’s annual report readability.
3. Materials and Method
3.1. Sample Selection
Along with job experience history and educational history, BoardEx reports any social organizations of which a company director or C-suite executive may currently be or formerly have been a part. Examples of these social organizations include city councils, professional guilds/societies, university trust boards, hospital advisory boards, sports clubs like golf or soccer clubs, religious organizations like churches, mosques or temples, religious charities, etc. For this study, we focus on religious organizations in which a C-suite executive may currently be or formerly have been a part of. One challenge with identifying religious organizations is that not all organizations with religious names or religious-sounding names are religious in nature. For example, affiliations with Mount Sinai Hospital in Manhattan, New York or Southern Methodist University in Dallas, Texas, are unlikely to be religious. Thus, to overcome these unique data mining challenges associated with identifying precise religious organization names from BoardEx, we use the GuideStar directory of charities and nonprofit organizations offered by Candid.2 GuideStar includes a comprehensive list of religious organizations in the U.S. for adherents of five of the largest global faith groups: Buddhism, Christianity, Hinduism, Islam, and Judaism. First, we use GuideStar to hand-collect keyword terms from religious organization names. Next, we match those keyword terms from GuideStar with the names of social organizations in BoardEx using text-based machine learning algorithms to identify the closest matches to existing religious organizations listed on GuideStar. We then run multiple filtering commands on the raw dataset to take out various religious-neutral organizations like hospitals or universities and any potential religious-neutral locations with a religious name (e.g., “Saint Joseph Street.”). This additional data cleaning step helps us better identify, match, and keep BoardEx-listed organizations that are religious.
We merge our hand-collected data with various publicly available data sources. First, we merge the cleaned religious organization data with organization, committee-level data available on BoardEx by the DirectorID identifier. We filter the data to keep only C-suite executives such as the CEO, CFO, CIO, COO, and CTO. Data relating to the Bog Index (Bonsall et al. 2017), our primary measure of annual report readability, is obtained from Professor Brian P. Miller’s website.3 Data relating to other relevant 10-K controls, such as gross file size or the number of 10-K exhibits, are obtained from Loughran and McDonald’s summarized 10-K datasheet.4 Financial data are obtained from Compustat. Finally, the Bog Index data, Loughran and McDonald data, and financial data are merged with the BoardEx and religious organization data. Our final dataset spans the years 1999 through 2020. This sample period is used primarily because of BoardEx data limitations pre-2000, which have also been highlighted by other studies such as Ke et al. (2019).
Panels A and B of Table 1 report summary and descriptive statistics relating to our sample, whereas Panel C reports Pearson correlations. Panel A reports that the mean Bog Score is 85.175. This average Bog Score is similar to the average Bog Score of 81.63 reported by Bonsall et al. (2017), even though their study’s sample period spanned between 1994 to 2011. In addition, Panel A shows that 6.60 percent of the sample firm-year observations have some form of disclosed religiosity at the executive level. Panel B reports that 2717 firms and 6497 C-suite executives are covered. CEOs make up 5677 of the executives in our sample. Panel B reports that 513 religious C-suite executives, of which 457 are CEOs, have identifiable/disclosed religious affiliations. The number of religious CEOs is the same as is reported in Baxamusa and Jalal (2016). The Pearson correlations from Panel C show that readability has a significant positive correlation with executive-level religiosity, which offers some preliminary support to the subsequent findings in the study. Location-based religiosity has a significant positive correlation with disclosed executive-level religiosity, but the correlation coefficient is only 2.70 percent. This modest correlation coefficient indicates little overlap between executive-level measures of religiosity and community-based measures of religiosity. Further, this low correlation suggests that CEOs may act based on the core themes of their religious beliefs independent of their workplace community’s religiosity. Therefore, this finding underscores the importance of using individual-based religiosity measures that have been advocated by recent religiosity studies in accounting and finance (Baxamusa and Jalal 2016; Cai et al. 2019; Chen et al. 2022).
3.2. Methodology
We modify the Bog Index Scores to make the interpretation of our regression results easier. Traditional Bog Index Scores are created to be positive numbers, with higher Bog Score values indicating poor quality readability and vice versa. Thus, to make the interpretation of our regressions coefficient easier, we follow Cassell et al. (2019) and multiply the Bog Index Scores by negative one to create a new variable labeled Readability Level. Then, we estimate our baseline predictive analytic regression model using the following specification:
Readability Level = ß1 Disclosed C-suite Religiosity + ß2 Document Characteristic Controls + ß3 Firm-Level Controls + ß4 Location Religiosity + ε.
Disclosed C-suite Religiosity, in our baseline predictive analytic regression model, is an indicator variable equal to one if a firm is led by one or more religious C-suite executives in a given year, whose religious identities are publicly identifiable from the religious organization affiliation data in BoardEx. In our baseline predictive analytic regression model, the vector Document Characteristic Controls include Gross File Size, which is the natural log of the file size of the 10-K and No. of Exhibits in 10-K, representing the total number of exhibits within the 10-Ks. Both control variables are obtained from Loughran and McDonald’s summarized 10-K datasheet. Firm-level controls in our baseline predictive analytic regression model are determined based on the extant literature and include controls for asset tangibility, profitability, a loss indicator, firm size, market-to-book ratio, firm age, firm-risk measures including earnings and cash flow volatility, book leverage, a new debt issuance dummy, capital expenditures, and sales, general and administrative (SGA) expenses. All firm-level variables are scaled by total assets. Based on the extant-location literature on religiosity, we also add a control for county-level religiosity (i.e., Location Religiosity) to our regression models. We use county-level data collected by the American Religious Data Archive (ARDA) in their 2000 and 2010 surveys to construct this Location Religiosity measure and follow the approach employed by Hilary and Hui (2009). The Location Religiosity measure captures the proportion of religious people in a given county, regardless of their religion or sub-religious sect.
The first column of Table 2, Table 3, Table 4, Table 5 and Table 6 contains our baseline predictive analytic regression model. We modify our baseline predictive analytic regression model by adding industry and year fixed effects to the second column in Table 2, Table 3, Table 4, Table 5 and Table 6. Next, we run propensity-score matching and entropy-balanced models as robustness checks in Table 7 and Table 8. As a final robustness check, we conduct a CEO turnover test to address endogeneity issues surrounding CEO religiosity in Table 9.
4. Results
4.1. Overall Findings
Table 2 examines the effect of executive-level religiosity on readability for all C-suite executives. The coefficient on Disclosed Religiosity from the baseline predictive analytic regression (coefficient = 1.639; t-statistic = 6.434) indicates that religiosity at the top of an organization has a strong positive impact on the readability of a firm’s annual report. When industry and year-fixed effects are added to the baseline predictive analytic regression model in the second column, the finding from column one continues to hold, as evident from the strongly positive and statistically significant coefficient on Disclosed Religiosity (coefficient = 0.763; t-statistic = 3.126). This finding supports the intrinsic dimension viewpoint, which argues that religious individuals generally do not try to pursue self-serving needs and usually take actions that maximize the welfare of the entire community. Under this perspective, religiosity will be reflected in higher degrees of transparency because religious individuals transparently share information. The statistically significant positive relationship between readability and C-suite religiosity in both specifications provide strong support for the intrinsic dimension viewpoint of religiosity. The results also validate that the individual religiosity of C-suite executives matters for annual report readability. It is not solely the religiosity of a firm’s headquarters location which drives the impact of religion. Our findings offer incremental evidence that individual-level religiosity affects corporate decision-making beyond the community-level of religiosity of the locale of firm headquarters.
Table 2.
Impact of Executive Level Religiosity on Annual Report Readability.
| Variables | Readability Level | Readability Level |
|---|---|---|
| Disclosed Religiosity | 1.639 *** | 0.763 *** |
| (6.434) | (3.126) | |
| Gross File Size | −1.505 *** | −0.857 *** |
| (−24.254) | (−8.458) | |
| No. of Exhibits in 10-K | −0.065 *** | −0.080 *** |
| (−4.004) | (−5.344) | |
| Tangibility | 7.461 *** | 4.762 *** |
| (21.282) | (11.170) | |
| Profitability | 3.012 *** | 2.017 *** |
| (5.561) | (4.165) | |
| Loss Indicator | −1.738 *** | −1.404 *** |
| (−10.855) | (−9.587) | |
| Size | −0.323 *** | −0.532 *** |
| (−8.216) | (−12.435) | |
| MTB | 0.005 *** | 0.005 *** |
| (8.142) | (8.882) | |
| Firm Age | 0.064 *** | 0.067 *** |
| (16.315) | (17.147) | |
| Earnings Volatility | 0.145 *** | 0.125 *** |
| (4.602) | (3.370) | |
| Cash Flow Volatility | −0.332 *** | −0.212 ** |
| (−3.957) | (−2.267) | |
| Book Leverage | −0.025 * | −0.048 *** |
| (−1.724) | (−3.072) | |
| Debt Issue | 0.101 | −0.004 |
| (0.746) | (−0.036) | |
| Capital Expenditure | 1.282 | −0.519 |
| (0.885) | (−0.345) | |
| SGA Expense | 2.953 *** | 1.948 *** |
| (5.459) | (4.023) | |
| Location Religiosity | −0.051 | −0.120 |
| (−0.128) | (−0.300) | |
| Constant | −64.286 *** | −71.237 *** |
| (−64.155) | (−49.154) | |
| Industry FE | N | Y |
| Year FE | N | Y |
| Observations | 11,391 | 11,391 |
| Adj. R-squared | 0.220 | 0.361 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
In Table 3 and Table 4, we conduct two subsample analyses to examine which executive within the management team drives the main results in Table 2. Recent research from Sherin (2010), Lee et al. (2012), and Ke et al. (2019) suggests that CEOs have the largest influence on corporate decision-making relative to other C-suite executives. CEO religiosity, therefore, should matter more than the religiosity of other executives to influence the readability of annual reports. As such, the evidence provided in Table 3 and Table 4 suggests that this is the case. Table 3, which only includes CEOs, reports a highly significant and positive relationship between individual religiosity and readability (coefficient = 1.596, 0.734; t-statistic = 5.985, 2.873). Whereas Table 4, which only includes non-CEO executives, returns mixed and insignificant coefficients (coefficient = 0.160, −0.130; t-statistic = 0.206, −0.177). These findings provide validation to the aforementioned evidence documented in Sherin (2010), Lee et al. (2012) and Ke et al. (2019) and indicate that CEO characteristics drive our results.
Table 3.
Impact of CEO Religiosity on Annual Report Readability.
| Variables | Readability Level | Readability Level |
|---|---|---|
| Disclosed Religiosity | 1.596 *** | 0.734 *** |
| (5.985) | (2.873) | |
| Gross File Size | −1.527 *** | −0.861 *** |
| (−24.265) | (−8.435) | |
| No. of Exhibits in 10-K | −0.060 *** | −0.076 *** |
| (−3.641) | (−5.022) | |
| Tangibility | 7.446 *** | 4.684 *** |
| (20.861) | (10.916) | |
| Profitability | 3.010 *** | 1.976 *** |
| (5.464) | (4.053) | |
| Loss Indicator | −1.770 *** | −1.447 *** |
| (−10.889) | (−9.753) | |
| Size | −0.320 *** | −0.532 *** |
| (−8.021) | (−12.313) | |
| MTB | 0.005 *** | 0.005 *** |
| (7.778) | (8.602) | |
| Firm Age | 0.064 *** | 0.067 *** |
| (16.183) | (17.021) | |
| Earnings Volatility | 0.144 *** | 0.122 *** |
| (4.567) | (3.241) | |
| Cash Flow Volatility | −0.331 *** | −0.207 ** |
| (−3.946) | (−2.239) | |
| Book Leverage | −0.023 | −0.050 *** |
| (−1.631) | (−3.122) | |
| Debt Issue | 0.076 | −0.027 |
| (0.554) | (−0.217) | |
| Capital Expenditure | 1.249 | −0.484 |
| (0.840) | (−0.315) | |
| SGA Expense | 2.951 *** | 1.907 *** |
| (5.364) | (3.914) | |
| Location Religiosity | −0.135 | −0.126 |
| (−0.334) | (−0.311) | |
| Constant | −63.948 *** | −71.160 *** |
| (−62.834) | (−48.696) | |
| Industry FE | N | Y |
| Year FE | N | Y |
| Observations | 11,149 | 11,149 |
| Adj. R-squared | 0.220 | 0.363 |
T-statistics are reported in parenthesis. *, ** and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
Table 4.
Impact of Non-CEO Executive Religiosity on Annual Report Readability.
| Variables | Readability Level | Readability Level |
|---|---|---|
| Disclosed Religiosity | 0.160 | −0.130 |
| (0.206) | (−0.177) | |
| Gross File Size | −1.259 *** | −0.682 ** |
| (−6.739) | (−2.273) | |
| No. of Exhibits in 10-K | −0.139 *** | −0.132 ** |
| (−2.675) | (−2.580) | |
| Tangibility | 8.383 *** | 6.360 *** |
| (9.213) | (4.609) | |
| Profitability | 1.181 | 0.590 |
| (1.011) | (0.819) | |
| Loss Indicator | −2.796 *** | −1.567 *** |
| (−6.100) | (−3.699) | |
| Size | −0.367 *** | −0.690 *** |
| (−2.988) | (−5.034) | |
| MTB | 0.011 | −0.006 |
| (0.621) | (−0.425) | |
| Firm Age | 0.031 ** | 0.056 *** |
| (2.480) | (4.483) | |
| Earnings Volatility | −0.038 | 0.124 |
| (−0.449) | (1.500) | |
| Cash Flow Volatility | −1.748 * | −0.095 |
| (−1.853) | (−0.104) | |
| Book Leverage | −0.313 *** | −0.167 * |
| (−2.805) | (−1.727) | |
| Debt Issue | 0.003 | −0.067 |
| (0.008) | (−0.182) | |
| Capital Expenditure | −5.392 | 0.649 |
| (−1.314) | (0.209) | |
| SGA Expense | 1.658 | 0.803 |
| (1.453) | (1.125) | |
| Location Religiosity | −0.298 | −1.562 |
| (−0.229) | (−1.084) | |
| Constant | −63.600 *** | −69.806 *** |
| (−21.707) | (−16.968) | |
| Industry FE | N | Y |
| Year FE | N | Y |
| Observations | 1287 | 1284 |
| Adj. R-squared | 0.198 | 0.419 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
4.2. Subsample Analysis—Christians vs. Non-Christian Religious Groups
Prior studies examine the overall religiosity levels within a given county without regard to whether the effect is driven by the adherents of the majority religion or whether the effect is driven by all religious groups. The analyses reported in Table 5 and Table 6 seek to understand whether minority religious groups within the U.S. have as much influence on corporate decision-making as the religious affiliation of the majority. Table 5 compares the Christian C-suite executives with their non-religious counterparts and excludes the religious C-suite executives who belong to non-Christian religious denominations. The coefficient on Disclosed Religiosity is positive and significant (coefficient = 1.970; t-statistic = 6.443). This evidence suggests that firms with executives who are affiliated with Christian faiths produce annual reports that are more readable.
Table 5.
Impact of Executive Level Religiosity on Annual Report Readability—Christian Group.
| Variables | Readability Level | Readability Level |
|---|---|---|
| Disclosed Religiosity | 1.970 *** | 1.076 *** |
| (6.443) | (3.748) | |
| Gross File Size | −1.504 *** | −0.877 *** |
| (−24.021) | (−8.599) | |
| No. of Exhibits in 10-K | −0.066 *** | −0.080 *** |
| (−3.996) | (−5.306) | |
| Tangibility | 7.514 *** | 4.757 *** |
| (21.192) | (11.156) | |
| Profitability | 2.962 *** | 2.022 *** |
| (5.456) | (4.122) | |
| Loss Indicator | −1.763 *** | −1.414 *** |
| (−10.941) | (−9.562) | |
| Size | −0.338 *** | −0.542 *** |
| (−8.532) | (−12.652) | |
| MTB | 0.005 *** | 0.005 *** |
| (8.094) | (8.782) | |
| Firm Age | 0.064 *** | 0.068 *** |
| (16.367) | (17.508) | |
| Earnings Volatility | 0.144 *** | 0.124 *** |
| (4.570) | (3.333) | |
| Cash Flow Volatility | −0.332 *** | −0.211 ** |
| (−3.980) | (−2.254) | |
| Book Leverage | −0.027 * | −0.051 *** |
| (−1.885) | (−3.200) | |
| Debt Issue | 0.108 | 0.002 |
| (0.793) | (0.015) | |
| Capital Expenditure | 1.034 | −0.608 |
| (0.703) | (−0.398) | |
| SGA Expense | 2.903 *** | 1.953 *** |
| (5.355) | (3.982) | |
| Location Religiosity | 0.035 | 0.032 |
| (0.087) | (0.080) | |
| Constant | −64.240 *** | −71.015 *** |
| (−63.685) | (−48.799) | |
| Industry FE | N | Y |
| Year FE | N | Y |
| Observations | 11,223 | 11,223 |
| Adj. R-squared | 0.221 | 0.363 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
Table 6, conversely, compares non-Christian, but religious C-suite executives with their non-religious counterparts and excludes the religious C-suite executives who belong to the Christian faith. The evidence from Table 6 is mixed. First, the baseline predictive analytic regression model results suggest that executive-level religiosity arising from non-Christian religious groups has a weakly significant and positive influence on annual report readability (coefficient on Disclosed Religiosity = 0.744; t-statistic = 1.726). This evidence is consistent with the broader conjecture of the intrinsic viewpoint, relating to religious individuals being responsible and transparent in their decision-making and actions, due to the potential implications of their actions on their communities. However, when industry and year fixed effects are included, the coefficient on Disclosed Religiosity is no longer significant (coefficient = 0.758; t-statistic = 0.684). Consistent with Table 2, Table 3 and Table 4, no evidence is found in Table 5 and Table 6 relating to the opportunistic use of religion by C-suite executives as conjectured by the proponents of the extrinsic dimension viewpoint.
Table 6.
Impact of Executive Level Religiosity on Annual Report Readability—Non-Christian Religious Groups.
Table 6.
Impact of Executive Level Religiosity on Annual Report Readability—Non-Christian Religious Groups.
| Variables | Readability Level | Readability Level |
|---|---|---|
| Disclosed Religiosity | 0.744 * | 0.758 |
| (1.726) | (0.684) | |
| Gross File Size | −1.529 *** | −0.296 *** |
| (−24.203) | (−3.079) | |
| No. of Exhibits in 10-K | −0.056 *** | 0.005 |
| (−3.449) | (0.342) | |
| Tangibility | 7.443 *** | 0.460 |
| (20.880) | (0.451) | |
| Profitability | 2.954 *** | 0.051 |
| (5.452) | (0.212) | |
| Loss Indicator | −1.742 *** | −0.675 *** |
| (−10.744) | (−5.097) | |
| Size | −0.327 *** | −0.552 *** |
| (−8.231) | (−3.022) | |
| MTB | 0.005 *** | 0.005 ** |
| (8.150) | (2.525) | |
| Firm Age | 0.061 *** | −0.775 *** |
| (15.442) | (−8.844) | |
| Earnings Volatility | 0.146 *** | −0.072 |
| (4.668) | (−0.702) | |
| Cash Flow Volatility | −0.336 *** | −0.105 |
| (−4.045) | (−0.233) | |
| Book Leverage | −0.026 * | 0.025 |
| (−1.785) | (0.840) | |
| Debt Issue | 0.119 | −0.005 |
| (0.866) | (−0.049) | |
| Capital Expenditure | 1.327 | 1.155 |
| (0.908) | (0.844) | |
| SGA Expense | 2.895 *** | −0.033 |
| (5.350) | (−0.136) | |
| Location Religiosity | −0.123 | 1.490 |
| (−0.300) | (1.018) | |
| Constant | −63.835 *** | −56.228 *** |
| (−62.523) | (−16.764) | |
| Industry FE | N | Y |
| Year FE | N | Y |
| Observations | 10,969 | 10,969 |
| Adj. R-squared | 0.214 | 0.342 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
4.3. Robustness Tests
In Table 7 and Table 8, we use propensity-score matching (PSM) and entropy-balancing to validate our core finding that individual religiosity at the top level of management affects the overall quality of firm disclosure practices. For the PSM and entropy-balanced specifications, we match the religious executive-led firms to the non-religious executive-led firms based on all firm-level and document characteristic controls. After the matching procedure, we follow Baxamusa and Jalal (2016) and estimate firm-level fixed effect regressions for the PSM specifications. For the entropy-balanced specification, we use standard linear regressions. Consistent with our earlier findings the coefficient on Disclosed Religiosity is positive and highly significant for both PSM specifications in Table 7 and Table 8 (coefficient = 1.324, 1.321; t-statistic = 3.648, 3.813) and the entropy balanced specifications in Table 7 and Table 8 (coefficient = 1.540, 1.274; t-statistic = 6.124, 4.912). The PSM and entropy-balancing-based matching procedures result in stricter comparisons between firms led by executives whose religious affiliations are identifiable versus those led by executives whose religious affiliations are not disclosed. Hence, these findings reiterate and validate the importance of individual religiosity in firm disclosure practices.
Table 7.
Impact of Executive Level Religiosity on Annual Report Readability—PSM & Entropy Approach.
| PSM | Entropy Balancing | |
|---|---|---|
| Variables | Readability Level | Readability Level |
| Disclosed Religiosity | 1.324 *** | 1.540 *** |
| (3.648) | (6.124) | |
| Gross File Size | −1.632 *** | −1.586 *** |
| (−9.204) | (−13.147) | |
| No. of Exhibits in 10-K | −0.097 ** | −0.086 ** |
| (−2.234) | (−2.396) | |
| Tangibility | 7.482 *** | 7.387 *** |
| (7.375) | (10.367) | |
| Profitability | 3.249 ** | 3.595 *** |
| (2.098) | (5.833) | |
| Loss Indicator | −1.277 ** | −1.796 *** |
| (−2.136) | (−3.927) | |
| Size | −0.293 ** | −0.112 |
| (−2.007) | (−1.187) | |
| MTB | 0.152 | 0.055 |
| (0.810) | (0.431) | |
| Firm Age | 0.075 *** | 0.068 *** |
| (6.420) | (8.155) | |
| Earnings Volatility | −2.207 | −4.975 ** |
| (−0.720) | (−2.297) | |
| Cash Flow Volatility | −6.153 | −1.036 |
| (−1.167) | (−0.357) | |
| Book Leverage | −0.360 | −0.277 |
| (−0.384) | (−0.547) | |
| Debt Issue | 0.500 | 0.197 |
| (1.163) | (0.682) | |
| Capital Expenditure | 0.545 | 0.643 |
| (0.099) | (0.190) | |
| SGA Expense | 5.108 *** | 4.783 *** |
| (6.123) | (11.751) | |
| Location Religiosity | −1.299 | −0.492 |
| (−1.065) | (−0.575) | |
| Constant | −62.114 *** | −64.382 *** |
| (−22.538) | (−34.553) | |
| Observations | 1520 | 11,391 |
| R-squared | 0.246 | 0.259 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
Table 8.
CEO Religiosity and Readability—PSM & Entropy Approach.
| PSM | Entropy Balancing | |
|---|---|---|
| Variables | Readability Level | Readability Level |
| Disclosed Religiosity | 1.321 *** | 1.274 *** |
| (3.813) | (4.912) | |
| Gross File Size | −1.479 *** | −1.535 *** |
| (−8.511) | (−12.092) | |
| No. of Exhibits in 10-K | −0.151 *** | −0.103 *** |
| (−3.281) | (−2.918) | |
| Tangibility | 7.598 *** | 8.734 *** |
| (7.631) | (11.504) | |
| Profitability | 2.274 ** | 3.895 *** |
| (2.105) | (5.977) | |
| Loss Indicator | −1.338 ** | −1.724 *** |
| (−2.205) | (−3.440) | |
| Size | −0.075 | −0.184 * |
| (−0.502) | (−1.890) | |
| MTB | 0.215 | |
| (1.022) | ||
| Firm Age | 0.068 *** | 0.068 *** |
| (6.308) | (8.214) | |
| Earnings Volatility | −8.854 *** | −7.325 *** |
| (−3.149) | (−3.420) | |
| Cash Flow Volatility | −1.333 | 0.381 |
| (−0.221) | (0.098) | |
| Book Leverage | −0.970 | 0.031 |
| (−1.159) | (0.062) | |
| Debt Issue | 0.860 ** | 0.175 |
| (2.260) | (0.598) | |
| Capital Expenditure | −0.671 | −7.281 * |
| (−0.151) | (−1.667) | |
| SGA Expense | 4.529 *** | 4.960 *** |
| (7.031) | (12.029) | |
| Location Religiosity | −1.744 | −1.474 * |
| (−1.418) | (−1.682) | |
| Constant | −64.516 *** | −63.952 *** |
| (−23.381) | (−32.716) | |
| Observations | 1378 | 11,857 |
| R-squared | 0.264 | 0.250 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. All variables are defined in Appendix A.
In Table 9, we conduct a CEO turnover test to address endogeneity issues surrounding CEO religiosity. We specifically focus on CEO turnovers from a non-religious to a religious CEO relative to CEO turnovers from a non-religious CEO to another non-religious CEO to examine the impact of CEO religiosity on the readability of 10-Ks. For this analysis, we define the two years prior to the CEO turnover as the PRE period and the two years following the CEO turnover as the POST period. The interaction variable PRE × Disclosed Religiosity does not differ from zero (coefficient = −5.845; t-statistic = −0.892) but POST × Disclosed Religiosity is positive and significant at the 1% level in the post-CEO turnover period (coefficient = 2.265; t-statistic = 3.648). These results suggest that a change from a non-religious to a religious CEO has a strong positive impact on the quality of corporate disclosure practices as measured by 10-K readability.
Table 9.
Impact of CEO Religiosity on Annual Report Readability--Non-Religious to Religious CEO Turnover Sample.
Table 9.
Impact of CEO Religiosity on Annual Report Readability--Non-Religious to Religious CEO Turnover Sample.
| Variables | Readability Level |
|---|---|
| PRE × Disclosed Religiosity | −5.845 |
| (−0.892) | |
| POST × Disclosed Religiosity | 2.265 *** |
| (3.648) | |
| Gross File Size | −1.085 *** |
| (−6.671) | |
| No. of Exhibits in 10-K | −0.066 ** |
| (−2.497) | |
| Tangibility | 3.376 *** |
| (5.213) | |
| Profitability | 1.875 *** |
| (6.022) | |
| Loss Indicator | −0.948 *** |
| (−4.197) | |
| Size | −0.801 *** |
| (−12.243) | |
| MTB | −0.010 ** |
| (−1.967) | |
| Firm Age | 0.057 *** |
| (7.852) | |
| Earnings Volatility | 0.153 *** |
| (5.921) | |
| Cash Flow Volatility | −0.283 *** |
| (−3.742) | |
| Book Leverage | 0.061 |
| (0.361) | |
| Debt Issue | −0.008 |
| (−0.036) | |
| Capital Expenditure | 1.862 |
| (0.941) | |
| SGA Expense | 1.919 *** |
| (6.193) | |
| Location Religiosity | 0.261 |
| (0.446) | |
| Constant | −66.793 *** |
| (−29.621) | |
| Industry FE | Y |
| Year FE | Y |
| Observations | 4437 |
| Adj. R-squared | 0.383 |
T-statistics are reported in parenthesis. *, **, and *** indicates statistical significance at the 10%, 5%, and 1% levels, respectively. There are 60 transitions from a non-religious to religious CEO, and 1207 transitions from a non-religious to non-religious CEO in our sample. All variables are defined in Appendix A.
5. Conclusions
This study extends the literature highlighting the importance of examining individual measures of religiosity to understand how religion affects corporate decision-making. The study also extends the strand of literature connecting religiosity with the quality of corporate disclosure practices. We find that firms with religious managers have more readable annual reports. Subsample analysis finds that CEOs drive the effect of individual religiosity at the C-suite level.
Author Contributions
Conceptualization, T.N.; methodology, T.N. and A.E.; formal analysis, T.N.; writing—original draft preparation, T.N.; writing—review and editing, T.N., A.E., K.E.D. and D.M.F.; All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
The authors have declared that this research is based on publicly available data.
Conflicts of Interest
The authors declare no conflict of interest.
Appendix A
| Variable | Definition |
| Readability Level | Bog Index *(−1) following Cassell et al. (2019). |
| Disclosed Religiosity | A dummy variable equal to 1 if a firm has one or more religious C-suite executives. |
| Gross File Size | Natural Log of 10-K file size. |
| No. of Exhibits in 10-K | Number of the firm’s 10-K exhibits. |
| Tangibility | Net PPE/Total Assets. |
| Profitability | Operating Income Before Income Tax & Depreciation/Total Assets. |
| Loss Indicator | Dummy variable equals 1 if income before extraordinary items < 0. |
| Size | Natural Log of Total Assets. |
| MTB | Market Value of Assets/ Book Value of Assets. |
| Market Value of Assets | Sum of the market value of equity plus debt in current liabilities plus long-term debt plus the liquidation value of the preferred stock less the deferred taxes and the investment tax credit following Baxamusa and Jalal (2016). |
| Firm Age | Age of firm. |
| Earnings Volatility | Measured as the standard deviation of earnings before extraordinary items scaled by total assets in 5-year rolling windows with a minimum of 5 observations, and winsorized at percentiles 1 and 99. |
| Cash Flow Volatility | Measured as the standard deviation of cash flows from operations scaled by total assets in 5-year rolling windows with a minimum of 5 observations, and winsorized at percentiles 1 and 99. |
| Book Leverage | (Current Liabilities + Long-Term Debt)/Total Assets. |
| Debt Issue | A dummy variable that takes a value of 1 if long-term debt issuance less long-term debt reduction is more than 1% of the total assets of the firm. |
| Capital Expenditure | Capital Expenditures/Total Assets. |
| SGA Expense | Selling, General and Administrative Expenses/Total Assets. |
| Location Religiosity | The overall religiosity of a county per 1000 members of population as reported by ARDA. |
| PRE | Year t − 1 and t − 2 prior to the CEO turnover. |
| POST | Year t + 1 and t + 2 following the CEO turnover. |
| 1 | Evidence that firms led by religious CEOs engage in less opportunistic earnings management and generally tend to have higher financial reporting quality (Cai et al. 2019; Chen et al. 2022) further supports this conjecture. |
| 2 | The GuideStar directory can be found at: https://www.guidestar.org/NonprofitDirectory.aspx (accessed on 18 February 2022). |
| 3 | Brian Miller’s website is located at: https://host.kelley.iu.edu/bpm/activities/bogindex.html (accessed on 27 March 2022). |
| 4 | The Loughran and McDonald data is available at: https://sraf.nd.edu/sec-edgar-data/lm_10x_summaries/ (accessed on 27 March 2022). |
References
- Adhikari, Binay Kumar, and Anup Agrawal. 2016. Religion, gambling attitudes and corporate innovation. Journal of Corporate Finance 37: 229–48. [Google Scholar] [CrossRef]
- Al Rahahleh, Naseem, M. Ishaq Bhatti, and Faridah Najuna Misman. 2019. Developments in risk management in Islamic finance: A review. Journal of Risk and Financial Management 12: 37. [Google Scholar] [CrossRef] [Green Version]
- America’s Changing Religious Landscape. 2015. Available online: https://www.pewresearch.org/religion/2015/05/12/americas-changing-religious-landscape/ (accessed on 1 February 2022).
- Anderson, Gary M. 1988. Mr. Smith and the preachers: The economics of religion in the wealth of nations. Journal of Political Economy 96: 1066–88. [Google Scholar] [CrossRef]
- Aribi, Zakaria Ali, and Simon S. Gao. 2012. Narrative disclosure of corporate social responsibility in Islamic financial institutions. Managerial Auditing Journal 27: 199–222. [Google Scholar] [CrossRef]
- Baxamusa, Mufaddal, and Abu Jalal. 2016. CEO’s religious affiliation and managerial conservatism. Financial Management 45: 67–104. [Google Scholar] [CrossRef] [Green Version]
- Beit-Hallahmi, Benjamin, and Michael Argyle. 1997. The Psychology of Religious Behaviour, Belief and Experience. Abingdon-on-Thames: Taylor & Francis. [Google Scholar]
- Bhatti, Maria. 2019. Managing Shariah non-compliance risk via Islamic dispute resolution. Journal of Risk and Financial Management 13: 2. [Google Scholar] [CrossRef] [Green Version]
- Bonsall, Samuel B., IV, Andrew J. Leone, Brian P. Miller, and Kristina Rennekamp. 2017. A plain English measure of financial reporting readability. Journal of Accounting and Economics 63: 329–57. [Google Scholar] [CrossRef]
- Bott, Jürgen, and Udo Milkau. 2018. Risk culture and the role model of the honorable merchant. Journal of Risk and Financial Management 11: 40. [Google Scholar] [CrossRef] [Green Version]
- Cai, Jay, and Guifeng Shi. 2019. Do religious norms influence corporate debt financing? Journal of Business Ethics 157: 159–82. [Google Scholar] [CrossRef]
- Cai, Ye, Yongtae Kim, Siqi Li, and Carrie Pan. 2019. Tone at the top: CEOs’ religious beliefs and earnings management. Journal of Banking & Finance 106: 195–213. [Google Scholar]
- Cano-Rodríguez, Manuel, and Alonso Moreno. 2020. Religion and the Readability of Textual Financial Disclosures. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3619116 (accessed on 1 February 2022).
- Cassell, Cory A., Lauren M. Cunningham, and Ling Lei Lisic. 2019. The readability of company responses to SEC comment letters and SEC 10-K filing review outcomes. Review of Accounting Studies 24: 1252–76. [Google Scholar] [CrossRef]
- Chen, Hanwen, Henry He Huang, Gerald J. Lobo, and Chong Wang. 2016a. Religiosity and the cost of debt. Journal of Banking & Finance 70: 70–85. [Google Scholar]
- Chen, Huaizhi, Lauren Cohen, and Dong Lou. 2016b. Industry window dressing. The Review of Financial Studies 29: 3354–93. [Google Scholar] [CrossRef]
- Chen, Yu, Xiaoyan Chu, Jung Chul Park, and Jared S. Soileau. 2022. CEO religious university affiliation and financial reporting quality. Accounting & Finance 62: 417–68. [Google Scholar]
- Chourou, Lamia, Luo He, and Ligang Zhong. 2020. Does religiosity enhance the quality of management earnings forecasts? Journal of Business Finance & Accounting 47: 910–48. [Google Scholar]
- Du, Xingqiang, Wei Jian, Shaojuan Lai, Yingjie Du, and Hongmei Pei. 2015. Does religion mitigate earnings management? Evidence from China. Journal of Business Ethics 131: 699–749. [Google Scholar] [CrossRef]
- Dyreng, Scott D., William J. Mayew, and Christopher D. Williams. 2012. Religious social norms and corporate financial reporting. Journal of Business Finance & Accounting 39: 845–75. [Google Scholar]
- Elamer, Ahmed A., Collins G. Ntim, Hussein A. Abdou, and Chris Pyke. 2020. Sharia supervisory boards, governance structures and operational risk disclosures: Evidence from Islamic banks in MENA countries. Global Finance Journal 46: 100488. [Google Scholar] [CrossRef]
- Geyer, Anne L., and Roy F. Baumeister. 2005. Religion, Morality, and Self-Control: Values, Virtues, and Vices. In Handbook of the Psychology of Religion and Spirituality. New York: The Guilford Press, pp. 412–32. [Google Scholar]
- Gibbins, Michael, Alan Richardson, and John Waterhouse. 1990. The management of corporate financial disclosure: Opportunism, ritualism, policies, and processes. Journal of Accounting Research 28: 121–43. [Google Scholar] [CrossRef]
- Grullon, Gustavo, George Kanatas, and James Weston. 2009. Religion and Corporate (Mis) Behavior. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1472118 (accessed on 1 February 2022).
- Guiso, Luigi, Paola Sapienza, and Luigi Zingales. 2003. People’s opium? Religion and economic attitudes. Journal of Monetary Economics 50: 225–82. [Google Scholar] [CrossRef] [Green Version]
- Hardy, Sam A. 2006. Identity, reasoning, and emotion: An empirical comparison of three sources of moral motivation. Motivation and Emotion 30: 205–13. [Google Scholar] [CrossRef]
- He, Wen, and Maggie Rong Hu. 2016. Religion and bank loan terms. Journal of Banking & Finance 64: 205–15. [Google Scholar]
- Hilary, Gilles, and Kai Wai Hui. 2009. Does religion matter in corporate decision making in America? Journal of Financial Economics 93: 455–73. [Google Scholar] [CrossRef]
- Iannaccone, Laurence R. 1998. Introduction to the Economics of Religion. Journal of Economic Literature 36: 1465–95. [Google Scholar]
- Jiang, Feng, Kose John, C. Wei Li, and Yiming Qian. 2018. Earthly Reward to the Religious: Religiosity and the Costs of Public and Private Debt. Journal of Financial and Quantitative Analysis 53: 2131–60. [Google Scholar] [CrossRef]
- Johnson, Paul E. 1959. Psychology of religion. In Psychology and Religion. Nashville: Abington Press. [Google Scholar]
- Jung, Carl Gustav. 1960. Psychology and religion. In Psychology and Religion. New Haven: Yale University Press. [Google Scholar]
- Ke, Ruihao, Meng Li, Zhejia Ling, and Yuan Zhang. 2019. Social connections within executive teams and management forecasts. Management Science 65: 439–57. [Google Scholar] [CrossRef]
- Kennedy, Ellen J., and Leigh Lawton. 1998. Religiousness and business ethics. Journal of Business Ethics 17: 163–75. [Google Scholar] [CrossRef]
- Lee, Sam, Steven R. Matsunaga, and Chul W. Park. 2012. Management forecast accuracy and CEO turnover. The Accounting Review 87: 2095–122. [Google Scholar] [CrossRef]
- Leventis, Stergios, Emmanouil Dedoulis, and Omneya Abdelsalam. 2018. The impact of religiosity on audit pricing. Journal of Business Ethics 148: 53–78. [Google Scholar] [CrossRef] [Green Version]
- Li, Ning. 2008. Religion, opportunism, and international market entry via non-equity alliances or joint ventures. Journal of Business Ethics 80: 771–89. [Google Scholar] [CrossRef]
- Lizińska, Joanna, and Leszek Czapiewski. 2019. Is window-dressing around going public beneficial? Evidence from Poland. Journal of Risk and Financial Management 12: 18. [Google Scholar] [CrossRef] [Green Version]
- Loughran, Tim, and Bill McDonald. 2011. When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. The Journal of Finance 66: 35–65. [Google Scholar] [CrossRef]
- Loughran, Tim, and Bill McDonald. 2014. Measuring readability in financial disclosures. The Journal of Finance 69: 1643–71. [Google Scholar] [CrossRef]
- Loughran, Tim, and Bill McDonald. 2016. Textual analysis in accounting and finance: A survey. Journal of Accounting Research 54: 1187–230. [Google Scholar] [CrossRef]
- Lyons, Bruce, and Judith Mehta. 1997. Contracts, opportunism and trust: Self-interest and social orientation. Cambridge Journal of Economics 21: 239–57. [Google Scholar] [CrossRef]
- Ma, Lijun, Min Zhang, Jingyu Gao, and Tingting Ye. 2020. The effect of religion on accounting conservatism. European Accounting Review 29: 383–407. [Google Scholar] [CrossRef]
- Malmendier, Ulrike, and Geoffrey Tate. 2009. Superstar CEOs. The Quarterly Journal of Economics 124: 1593–638. [Google Scholar] [CrossRef]
- McGuire, Sean T., Thomas C. Omer, and Nathan Y. Sharp. 2012. The impact of religion on financial reporting irregularities. The Accounting Review 87: 645–73. [Google Scholar] [CrossRef]
- Middleton, Russell, and Snell Putney. 1962. Religion, normative standards, and behavior. Sociometry 25: 141–52. [Google Scholar] [CrossRef]
- Milkau, Udo. 2017. Risk culture during the last 2000 years—From an aleatory society to the illusion of risk control. International Journal of Financial Studies 5: 31. [Google Scholar] [CrossRef] [Green Version]
- Oh, Frederick Dongchuhl, and Donglim Shin. 2020. Religion and Corporate Disclosure Quality. Hitotsubashi Journal of Economics 61: 20–37. [Google Scholar]
- Omer, Thomas C., Nathan Y. Sharp, and Dechun Wang. 2018. The impact of religion on the going concern reporting decisions of local audit offices. Journal of Business Ethics 149: 811–31. [Google Scholar] [CrossRef]
- Pargament, Kenneth I., Gina M. Magyar-Russell, and Nichole A. Murray-Swank. 2005. The sacred and the search for significance: Religion as a unique process. Journal of Social Issues 61: 665–87. [Google Scholar] [CrossRef]
- Rawwas, Mohammed YA, Ziad Swaidan, and Jamal Al-Khatib. 2006. Does religion matter? A comparison study of the ethical beliefs of marketing students of religious and secular universities in Japan. Journal of Business Ethics 65: 69–86. [Google Scholar] [CrossRef]
- Roccas, Sonia. 2005. Religion and value systems. Journal of Social Issues 61: 747–59. [Google Scholar] [CrossRef]
- Salsman, John M., Tamara L. Brown, Emily H. Brechting, and Charles R. Carlson. 2005. The link between religion and spirituality and psychological adjustment: The mediating role of optimism and social support. Personality and Social Psychology Bulletin 31: 522–35. [Google Scholar] [CrossRef] [PubMed]
- Sherin, Keith. 2010. Financial planning and investor communications at GE (with a look at why we ended earnings guidance). Journal of Applied Corporate Finance 22: 8–17. [Google Scholar] [CrossRef]
- Smith, Adam. 1776. An Inquiry into the Nature and Causes of the Wealth of Nations. London: Printed for W. Strahan and T. Cadell. [Google Scholar]
- Stulz, Rene M., and Rohan Williamson. 2003. Culture, openness, and finance. Journal of Financial Economics 70: 313–49. [Google Scholar] [CrossRef]
- Sunstein, Cass R. 1996. Social norms and social roles. Columbia Law Review 96: 903. [Google Scholar] [CrossRef]
- Vitell, Scott J., Mark N. Bing, H. Kristl Davison, Anthony P. Ammeter, Bart L. Garner, and Milorad M. Novicevic. 2009. Religiosity and moral identity: The mediating role of self-control. Journal of Business Ethics 88: 601–13. [Google Scholar] [CrossRef]
- Wulff, David M. 1991. Psychology of Religion: Classic and Contemporary Views. Hoboken: John Wiley & Sons. [Google Scholar]
Table 1.
Summary statistics, sample selection and Pearson correlations.
| Panel A: Summary Statistics | ||||||||||||||||
| Mean | SD | p25 | Median | p75 | ||||||||||||
| Bog Score | 85.175 | 7.567 | 80 | 85 | 90 | |||||||||||
| Disclosed Religiosity | 0.066 | 0.249 | 0.000 | 0.000 | 0.000 | |||||||||||
| Gross File Size | 14.912 | 1.444 | 13.874 | 14.638 | 16.293 | |||||||||||
| No. of Exhibits in 10-K | 10.518 | 5.040 | 7 | 10 | 13 | |||||||||||
| Tangibility | 0.269 | 0.238 | 0.085 | 0.188 | 0.394 | |||||||||||
| Profitability | −0.060 | 5.634 | 0.060 | 0.115 | 0.167 | |||||||||||
| Loss Indicator | 0.299 | 0.458 | 0 | 0 | 1 | |||||||||||
| Size | 6.741 | 2.238 | 5.284 | 6.882 | 8.197 | |||||||||||
| MTB | 3.064 | 65.057 | 0.840 | 1.225 | 1.980 | |||||||||||
| Firm Age | 26.490 | 17.659 | 12 | 22 | 40 | |||||||||||
| Earnings Volatility | 0.265 | 4.651 | 0.018 | 0.037 | 0.093 | |||||||||||
| Cash Flow Volatility | 0.134 | 1.956 | 0.024 | 0.041 | 0.077 | |||||||||||
| Book Leverage | 0.375 | 7.043 | 0.058 | 0.217 | 0.366 | |||||||||||
| Debt Issue | 0.320 | 0.467 | 0 | 0 | 1 | |||||||||||
| Capital Expenditure | 0.050 | 0.061 | 0.016 | 0.032 | 0.060 | |||||||||||
| SGA Expense | 0.419 | 5.839 | 0.103 | 0.205 | 0.368 | |||||||||||
| Location Religiosity | 0.594 | 0.159 | 0.470 | 0.587 | 0.685 | |||||||||||
| Panel B: Sample Selection | ||||||||||||||||
| Details | Count | |||||||||||||||
| C-Suite Executives with Disclosed Religiosity | 513 | |||||||||||||||
| Total No. of C-Suite Executives | 6497 | |||||||||||||||
| CEOs with Disclosed Religiosity | 457 | |||||||||||||||
| Total No. of CEOs | 5677 | |||||||||||||||
| Total No. of Firms | 2717 | |||||||||||||||
| Panel C: Pearson Correlations | ||||||||||||||||
| Variables | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | (12) | (13) | (14) | (15) | (16) |
| Readability (1) | ||||||||||||||||
| Disclosed Religiosity (2) | 0.069 | |||||||||||||||
| Gross File Size | −0.315 | −0.011 | ||||||||||||||
| No. of Exhibits in 10-K (3) | −0.215 | 0.004 | 0.652 | |||||||||||||
| Tangibility (4) | 0.226 | 0.015 | 0.038 | 0.069 | ||||||||||||
| Profitability (5) | −0.006 | 0.009 | 0.013 | 0.015 | 0.028 | |||||||||||
| Loss Indicator (6) | −0.164 | −0.061 | −0.003 | −0.021 | −0.047 | −0.052 | ||||||||||
| Size (7) | −0.064 | 0.051 | 0.308 | 0.332 | 0.181 | 0.136 | −0.340 | |||||||||
| MTB (8) | 0.023 | −0.006 | −0.012 | −0.006 | −0.025 | −0.693 | 0.030 | −0.112 | ||||||||
| Firm Age (9) | 0.081 | 0.058 | 0.229 | 0.211 | 0.083 | 0.014 | −0.217 | 0.397 | −0.009 | |||||||
| Earnings Volatility (10) | 0.003 | −0.012 | −0.009 | −0.015 | −0.037 | −0.755 | 0.062 | −0.143 | 0.515 | −0.027 | ||||||
| Cash Flow Volatility (11) | −0.010 | −0.011 | −0.012 | −0.020 | −0.040 | −0.633 | 0.063 | −0.128 | 0.422 | −0.034 | 0.930 | |||||
| Book Leverage (12) | 0.009 | −0.003 | 0.011 | 0.008 | −0.010 | −0.576 | 0.022 | −0.078 | 0.232 | −0.003 | 0.572 | 0.621 | ||||
| Debt Issue (13) | 0.008 | 0.013 | 0.061 | 0.071 | 0.113 | 0.016 | −0.027 | 0.169 | −0.015 | 0.061 | −0.018 | −0.018 | −0.001 | |||
| Capital Expenditure (14) | 0.164 | −0.009 | −0.026 | −0.007 | 0.597 | 0.017 | −0.049 | 0.039 | −0.014 | −0.049 | −0.017 | −0.018 | −0.010 | 0.137 | ||
| SGA Expense (15) | 0.018 | −0.007 | −0.021 | −0.022 | −0.036 | −0.998 | 0.040 | −0.146 | 0.692 | −0.015 | 0.757 | 0.637 | 0.576 | −0.020 | −0.020 | |
| Location Religiosity (16) | 0.111 | 0.027 | −0.351 | −0.195 | −0.047 | 0.014 | −0.049 | 0.021 | −0.007 | 0.047 | −0.017 | −0.017 | −0.005 | −0.003 | −0.018 | −0.011 |
All variables are defined in Appendix A. Correlation coefficients in bold are significant at the 1% level.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Nazrul, T.; Esplin, A.; Dow, K.E.; Folsom, D.M. Religiosity at the Top and Annual Report Readability. J. Risk Financial Manag. 2022, 15, 485. https://doi.org/10.3390/jrfm15100485
AMA Style
Nazrul T, Esplin A, Dow KE, Folsom DM. Religiosity at the Top and Annual Report Readability. Journal of Risk and Financial Management. 2022; 15(10):485. https://doi.org/10.3390/jrfm15100485
Chicago/Turabian StyleNazrul, Toufiq, Adam Esplin, Kevin E. Dow, and David M. Folsom. 2022. "Religiosity at the Top and Annual Report Readability" Journal of Risk and Financial Management 15, no. 10: 485. https://doi.org/10.3390/jrfm15100485
