Received: 13 October 2018 Revised: 13 December 2019 Accepted: 18 June 2020

DOI: 10.1002/ijfe.1994

RESEARCH ARTICLE

Religiosity and financial distress in U.S. firms

Ines Gharbi1 | Mounira Hamed-Sidhom1 | Khaled Hussainey2 |

Janet Ganouati3,4

1
Faculty of Economic Sciences and
Management of Tunis, UR17ES07 FCF
Abstract
Research Unit, University of Tunis El In our paper, we test the global impact of religiosity on firm's durability. Given
Manar, Tunis, Tunisia that religious firms are more ethics and take less risk, they avoid the costs of
2
Faculty of Business and Law, University
misconduct, and they benefit from the good reputation and the excellent rela-
of Portsmouth, Portsmouth, UK
3
BADEM Lab, Tunis Business School,
tionship with their stakeholders. So, we predict that higher degrees of religios-
Université de Tunis, Tunis, Tunisia ity can reduce the financial distress. According to this prediction, we detect
4
LEM CNRS UMR 9221, Université de that corporates headquarters situated in more religious U.S. counties are prob-
Lille, Faculté des Sciences Économiques ably less to suffer from financial problems. We also note that this negative rela-
et Sociales, Université des Sciences et
Technologies de Lille, Lille, France tion becomes stronger during the crisis period. We conclude that the lack of
religiosity is a significant cause of the financial difficulty.
Correspondence
Janet Ganouati, BADEM Lab, Tunis KEYWORDS
Business School, Université de Tunis,
Ethics, financial distress, religiosity, risk aversion
n
65 Bir El Kassaa 2059, Tunis Tunisia.
Email: [email protected]

1 | INTRODUCTION finance. They note that religion influences finance across

at least three ways. First, the ethics and values depend on
The financial crisis is a spiritual crisis. In fact, this crisis the religion of the country. For example, charging inter-
is a values crisis. Many religious leaders advise people to est is different from country to another. Second, the insti-
remove their money from the big banks that have proved tutions depend on the religion. Indeed, the legal system
an enormous lack of morals (Wallis, 2010). $8.7 trillion is affected by culture. Finally, culture influences how
investor dollars managed employing faith-based or other resources are distributed in an economy. Indeed, the reli-
screening methods. The number of faith-based advisers is gion which encourages the expenditure of money in
rising. They advise their clients to avoid investing in “sin churches or guns decreases the resources using for the
stocks” like tobacco and gambling. Indeed, incorporating investment in production. Stulz and Williamson (2003)
religious values into financial investment reduce debt; provide evidence that differences in culture are essential
have a positive social impact and a positive financial to understanding why investor protection differs across
return. Tergesen, 2017. countries. In fact, Catholic countries have significantly
Moreover, there is a significant evolution in the rela- weaker creditor rights than other countries. The role
tions between the corporate world and faith. Sister Pat- played by culture in economic life has been widely dis-
ricia Daly is a Dominican nun. She engages in putting cussed. Rich literature found that the all differences
pressure on the firms to be more socially responsible. The between countries can be explained by culture
General Electric executive said that for many years many (Greif, 1994; Lal, 1999; Landes, 2000; North, 1990). All
people avoid meeting Sister Patricia Daly, but now they these researches examine the influence of culture on eco-
look for her (Gibson, 2017). nomic growth across countries.
The earlier studies have used religion as a proxy for Then, papers used the local religiosity of the county
culture across countries. Stulz and Williamson (2003) in the United States as a proxy for corporate culture
tried to find a direct association between culture and because it is quite easy to measure (Jiang, John, Li, &

Int J Fin Econ. 2020;1–14. wileyonlinelibrary.com/journal/ijfe © 2020 John Wiley & Sons, Ltd. 1
2 GHARBI ET AL.

Qian, 2018). Ahmed, Loh, and Zairi (1999) defined cor- crisis. Moreover, until now, media coverage accuses the
porate culture as “the pattern of arrangement, material human nature. (Sonad, 2018).
or behaviour which has been adopted by a society Giving that religiosity decreases the risk preferences
(corporation, group, or team) as the accepted way of of managers (He & Hu, 2016) and in the other hand, this
solving problems.” However, how religious attitudes risk preference promotes the corporate's risk-taking
affect corporate behaviour is not well assumed. behaviour (Adhikari & Agrawal, 2016) it is fascinating to
Lagace (2001) consider the combination of personal see if religious firms have been able to overcome finan-
religious values with organizational behaviour is one cial problems in the period of crisis.
of the areas of research that are not sufficiently inves- Finally, according to the analyses conducted by
tigated. Many studies argue that firms located in more Moody's Analytics about the states in recession1 and the
religious areas are less likely to be implicated in survey of Gallup2 concerning the most and the least reli-
accounting fraud, tax avoidance, and stock price gious states in the United States in 2008, three remarks
crashes (Boone, Khurana, & Raman, 2013; Callen & can be detected. First, the impact of the financial crisis
Fang, 2015; Grullon, Kanatas, & Weston, 2010; depends on the geographic locations. So, some states had a
McGuire, Omer, & Sharp, 2012). tolerable expansion. Nevertheless, others were in a reces-
This paper expands on this area of research by explor- sion. Then, the religiosity in the U.S. context varies from
ing how local religious norms are a kind of social influ- state to state. For this reason, many papers treated the
ence affecting the firm's durability. Specifically, we impact of religiosity on corporate behaviour in the
examine whether religious social norms have an impact U.S. context. These papers overcome the problem linked
on financial distress in the United States. to the international context. In fact, studying the effect of
Three factors motivate this study. religion on a firm's decisions like earning manipulations
First is the lack of research on the effect of religiosity in the international context (Kanagaretnam et al. 2014)
on the financial distress. The earlier research treated the will create a significant problem. In this case, it is compli-
relation between individual religiosity and risk aversion cated to separate the effect of the country's legal and insti-
at the individual level such as Ahmad (1973) and tutional characteristics from the impact of religiosity. So,
Rokeach (1968). At a later stage, many papers discussed studying the effect of religiosity in the U.S. context allows
how individual religiosity affects corporate behaviour. us to control the relationship. Moreover, it is a very rich
Hilary and Hui (2009) studied the relation between religi- context for testing our hypotheses, and it contains
osity and making decision in U.S. firms. Adhikari and 11 nations with different cultures (Speiser, 2015).
Agrawal (2016) examined whether the risk preferences of Third, in 2008, the most religious states had a tolera-
managers influence a bank's risk-taking behaviour. ble expansion such as Texas with 74% of religious resi-
Grullon et al. (2010) treat the relation between religiosity dents, Alabama with 82% of religious residents and
and excessive compensation. Other papers report a con- Oklahoma with 75% of religious residents. However, the
nection between religiosity and the firm's decisions such least religious states were in a recession like Nevada with
as leverage, tax avoidance, accounting manipulation, but 54% religious residents and Massachusetts with 48% reli-
studying the consequences of these behaviours on the gious residents. So, it is very interesting to study the
financial health of the company is not very explored impact of religiosity on a firm's financial distress.
except the paper of Callen and Fang (2015) which investi- Our paper extends the literature by providing new
gates the relation between religiosity and future stock evidence regarding the role of religiosity in the economy.
price crash risk. If religiosity has an impact on the finan- First, our study on religiosity complements the papers
cial health of the firms, it would be beneficial to dealing with the importance of psychological skills in
standard-setters, investors, and regulators to take into corporate decisions such as CEO overconfidence (Leng,
consideration this factor (Barro & McCleary, 2003; Guiso, Trzeciakiewicz, & Ozkan, 2018) and CEO narcissism
Sapienza, & Zingales, 2006). (Rijsenbilt & Commandeur, 2013). Second, studying the
Second, according to a survey conducted by effect of religiosity on corporate financial health consti-
PricewaterhouseCoopers and the Economist Intelligence tutes a global view of the importance of social norms and
Unit in 2008, 73% of participants find that culture and complements the previous papers which take a single
excessive risk taking are the most important causes of impact of religiosity like Hilary and Hui (2009), Adhikari
financial crisis. Furthermore, in a testimony to the and Agrawal (2016), Li, Wang, and Wang (2017).
U.S. House of Representatives, Lo (2008) says “…the ulti- We note that firms located in more religious regions
mate origin of the crisis may be human behavior….” are less likely to have financial distress; according to the
Indeed, it seems that human behaviours could explain view that local religiosity induces a risk-averse corporate
the firms' risk-taking behaviour during the financial culture and a high degree of ethics. In particular, local
GHARBI ET AL. 3

religiosity negatively influences financial distress mea- Social norms are reinforced by the social proof heuris-
sured by the inverse of Altman Z-score. This negative tic (Cialdini, 1993). This psychological phenomenon stip-
relation becomes stronger in the financial crisis of 2008. ulates that individuals follow the surrounding people
This finding is coherent with the view that firms with because they believe that the dominant group has more
religious culture are less susceptible to crisis. Our results knowledge about the correct behaviour (Cialdini, 1993).
still robust when we use the revised Atman Z" score as a Therefore, geographic locations with strong religious
proxy for financial distress. beliefs increase the likelihood that managers are affected
This paper is organized as follows. Section 2 reviews by religious social norms (Kennedy & Lawton, 1998).
the social norm theory and the literature on the associa- In the beginning, researches in finance and account-
tion between religiosity and financial distress. Section 3 ing ignored the effect of human behaviour on the firm
presents the data and variables. Empirical results are pro- policies. After that, growing literature suggests that the
vided in Section 4. Section 5 reports robustness checks. human aspect was a missing link in the corporate mak-
Finally, Section 6 concludes. ing decisions (Bertrand & Schoar, 2003; Malmendier,
Tate, & Yan, 2011). Indeed, the first paper investigates
the effect of local religiosity on firm profitability and risk
2 | L I T E R A T U R E R EVI E W A N D aversion was the work of Hilary and Hui (2009). Then,
HYPOTHESES DEVELOPMENT researches were interested in the impact of religiosity on
the financial reporting quality (Dyreng et al., 2012; Li
There is two stream of research explain the relationship et al., 2017; McGuire et al., 2012), on the excessive com-
between religion and economic. The first one is macroeco- pensation (Grullon et al., 2010), on the risk aversion
nomic. It relates religion to economic growth and higher (Gao, Wang, & Zhao, 2017; He & Hu, 2016), on the crash
per capita income (Barro & McCleary, 2003; Guiso, risk (Callen & Fang, 2015), on the credit rating and debt
Sapienza, & Zingales, 2003), explains how a country's prin- cost (Jiang et al., 2018).
cipal religion predicts the cross-country variations in All these researches highlight the importance of reli-
investor protection (Stulz & Williamson, 2003) and docu- giosity on the corporate policies and decisions. This effect
ments that sharing the same religion can promote the for- can influence the durability of the firm. Indeed, the liter-
eign direct investments between two countries (Guiso, ature demonstrates that some traits of CEO affect the
Sapienza, & Zingales, 2009). This stream of literature financial distress: overconfidence increases the financial
explains the role played by religion in the economic failure (Leng et al., 2018), CEO narcissism has a positive
growth. The second stream describes how individual relationship with accounting manipulation (Rijsenbilt &
behaviours conducted by beliefs and ideologies affect Commandeur, 2013).
financial behaviours (Hirshleifer, 2014). In this case, we The literature has focused on two characteristics
talk about religiosity and not religion. In fact, the level of about firms located in regions with high religiosity (Jiang
religiosity in the United States counties explains the corpo- et al., 2018): ethics (Duarte, Siegel, & Young, 2012; Li
rate behaviours in U.S. firms: risk exposures (Hilary & et al., 2017) and risk aversion (He & Hu, 2016).
Hui, 2009), quality of financial reporting (Dyreng, May-
ew, & Williams, 2012; McGuire et al., 2012; Li et al., 2017),
and unethical misconducts (Grullon et al., 2010). 3 | ETHICS
Departing from the idea that corporate decisions are
made by individuals, not firms (Hilary & Hui, 2009), the Many papers find that religious firms have a lower
social norm theory predicts that the dominant norms and accounting manipulation, a lower tax avoidance, a higher
beliefs in the geographic location of the firms influence accounting conservatism (Li et al., 2017). Moreover,
the decision made by managers. In other words, religious Omer, Sharp, and Wang (2016) provide evidence that
standards will affect the behaviour of managers even they religiosity influences auditor opinions.
may or may not be religious because these latter live and This ethical behaviour is essential for many reasons.
operate in an environment which social rules of the local First, it can prevent the damage related to misconduct
population constitute an essential element (Cialdini & behaviour. Indeed, there is a negative relation between
Goldstein, 2004; Sunstein, 1996). earning management and firm performance (Fairfield,
Indeed, individuals desire to comply with the con- Whisenant, & Yohn, 2003) and a negative relation
ducts of the others because they want to avoid the between tax avoidance and stock price (Hanlon &
expenses or penalties associated with rejecting the stan- Slemrod, 2009). Regarding that firms located in more reli-
dards or beliefs that are considered admissible or suitable gious counties are less likely to experience accounting
for the local people (Sunstein, 1996). fraud due to the importance of social norms in the
4 GHARBI ET AL.

religious community (Dyreng et al., 2012; Grullon 4 | RISK AVERSION

et al., 2010), they are more likely to avoid the cost of
misconduct. Another trait of religious firms is the risk aversion. Begin-
Second, ethical behaviour can minimize the agency ning with the paper of Hilary and Hui (2009), these
costs which can be classified into two kinds. The first authors point out that religious firms invest less than no
kind of agency cost is related to the divergence between religious firms. Indeed, religious firms do not prefer risk.
the agent's interest and the principal's interest. Giving However, Kumar et al. (2011) mention that the gambling
that ethics provide members the ability to distinguish attitude depends on the religion. For instance, Protes-
between ethical and unethical experience, religious tants have strong opposition. However, Roman Catholics
firms do not consider earning manipulation as ethical have a moderate view for gambling. Adhikari and
behaviour (Weaver & Agle, 2002). Moreover, Merchant Agrawal (2016) study the impact of religiosity on the vari-
and Rockness (1994) find that religiosity and education ation of bank risk taking. They find that banks located in
affect the ethical judgements about earnings manage- more religious regions have a high risk aversion. Indeed,
ment and Longenecker, McKinney, and Moore (2004) they increase their assets carefully, they invest less in
consider that religious beliefs influence the accounting risky products, and they do not encourage their managers
choices of CEO such as religious CEO prefers to use to take many risks. So, they are less exposed to crises
real earning management than accruals manipulations than the no religious banks. In the same order of idea,
(Hilary & Hui, 2009). Furthermore, Du (2012) finds a He and Hu (2016) argue that banks give favourable loans
negative relationship between religion and principal- conditions for firms located in religious areas because
agent agency cost. He concludes that religion consti- they are less vulnerable to financial problems.
tutes a mechanism curbing the agency cost. The second The theoretical framework of our research is summa-
kind of agency costs is related to mechanisms using to rized in the Figure 1. According to social norm theory,
align the interests of the manager with those of the managers are influenced by social norms in more reli-
shareholders. Hilary and Hui (2009) suggest that the gious areas. Many papers find that religious firms have
impact of religiosity on the financial reporting irregu- higher performance, lower accounting manipulation,
larities is widespread especially when we have weak lower tax avoidance, and favourable loan conditions. For
corporate governance. In this case, social norms play these reasons, they maintain a good relationship with
the role of control and substitute the governance mech- their shareholders and their stakeholders such as lenders,
anisms (Du, 2012). To sum up, religious firms are less employees, government. In the same order of idea, the
likely to have financial problems since religiosity can misconduct of Enron consists of using earning manage-
reduce the agency cost. ment to create values for own interests, having a high
Third, the ethical behaviour can maintain and level of tax avoidance and less accounting transparency
improve the firm's reputation. So, corporates built and (Tonge, Greer, & Lawton, 2003). So, this corporate had
leverage a good relationship with their stakeholders. lost the trust of their stakeholders (McAfee, 2004).
He and Hu (2016) find that firms with high religious Given above, we expect that religious firms suffer less
values have favourable loan conditions. Furthermore, from financial problems. Put differently; there is a nega-
the increase of the level of religiosity reduces about 8.9 tive relationship between religiosity and financial
basis points of the level of loan spread. Indeed, religi- distress.
osity is based on trust, and trust is elementary for the
lending decisions (Guiso et al., 2003). Furthermore, H1 Firms in lowly religious regions have more financial
the reliability constitutes an essential condition for distress than firms in highly religious areas.
loans (Duarte et al., 2012). In the same order of idea,
Guiso et al. (2015) find that employees work hard if After the last crisis, many papers seek to understand
they consider the CEO as trustworthy and honest. So, the reasons for the financial crisis. Among the causes,
religious firms are more socially responsible than corporate culture has received much attention. It seems a
others firms (Kim, Yadav, Han, & Yang, 2018). For this crucial factor.
reason, they can maintain a good reputation and avoid Kanagaretnam, Lim, and Lobo (2014) try to test the
some costs. relation between some traits of culture and having finan-
Finally, the ethical behaviour increases the sales of cial problems during the crisis. They find that banks in
the firms. In fact, Hilary and Hui (2009) find that reli- low individualism and high uncertainty avoidance cul-
gious firms have a good performance. Furthermore, firms tures are less likely to have financial troubles during the
headquartered in areas of high social faith tend to have financial crisis. Indeed, people with high uncertainty
smaller crash risks (Li et al., 2017). avoidance do not appreciate the ambiguity and the
GHARBI ET AL. 5

F I G U R E 1 Theoretical
framework of our research

uncertainty. For this reason, they take less risk and they and 2010. Following Hilary and Hui (2009), we obtain
prefer a lower return with a known risk. In the same estimates for the intermediate years by linearly interpo-
way, Adhikari and Agrawal (2016) note the negative rela- lating the decennial data.
tionship between religiosity and bank risk taking The locations of firms' headquarters are obtained
becomes more intense during the crisis of 1998 and from COMPUSTAT to match firm and county-level data.
2007–08. Banks in more religious areas are more attached However, the number of observations is very low. We
to their culture and behaviours during the crisis, and they determine the location of missing firms manually by
learn from the financial troubles. matching postal code and state.i We exclude all financial
services (two-digit SIC codes between 60 and 69) from
H2 The negative relation between financial distress and the sample because high leverage in financial firms does
religiosity becomes stronger during the crisis. not have the same meaning in the others firms (Fama &
French, 1992). Our sample period is from 1974 through
2010. Our analysis includes 8,333 firms and 78,317 firm-
5 | S AM PL E A ND M E AS UR E O F years of observations.
R E L I G I O S I T Y AN D F I N A N C I A L
DISTRESS
5.1 | Measuring religiosity
To test these hypotheses, we use COMPUSTAT North
America. We focus only on U.S. firms for the same reason Following Hilary and Hui (2009), we estimate the religios-
presented by Hilary and Hui (2009) and Adhikari and ity of a firm by the ratio of religious members to the popu-
Agrawal (2016). Basically, religiosity differs from county to lation of the county where the firm is headquartered. Two
another in the United States. So, we can isolate the impact reasons can explain the efficacy of this ratio. First, in
of a country's legal and institutional characteristics from United States, employees are likely to work in their local
the effect of religiosity. We consider a firm's location as the communities. So, firms located in religious areas have a
location of its headquarters. Pirinsky and Wang (2006) great proportion of religious employees. Second, according
suggest that headquarters are usually close to a firm's main to social norm theory, people tend to follow the dominant
activities. Also, following prior studies, we admit a conta- beliefs and the behaviour of people around them.
gion effect of local norms. Individuals are affected by the
dominant local culture even if they do not share it.
Data on religiosity took from the Churches and 5.2 | Measuring financial distress
Church Membership files of the American Religion Data
Archive (ARDA) website, which has county-level religion The measure of financial trouble is based on the Altman
statistics on Judeo-Christian bodies every 10 years. Religi- Z-score (Altman, 1968). In fact, financial ratios have a
osity data is available for five years 1971, 1980, 1990, 2000 high probability to predict corporate failure.
6 GHARBI ET AL.

Although given many critics related to the methodol- 6 | SAMPLE DESCRIPTION A ND

ogy, this accounting score is still used by literature UNIVARIATE ANALYS IS
(Bhaskar, Krishnan, & YU, 2017; Chen, De Cesari, Hill, &
Ozkan, 2018; Donker, Ng, & Shao, 2018; Richardson, 6.1 | Descriptive statistics
Lanis, & Taylor, 2015). In fact, the results of this score
should be interpreted carefully when the research is First, we describe our sample across industries. Second,
about non-U.S. firms and non-manufacturing firms. To we present the summary statistics for dependent, inde-
deal with the last problem, we conduct a robustness test pendents and control variables.
to check the effectiveness of this score. Table 1 envisages the industrial distribution of our
Basically, the Altman Z-score is more able to predict sample based on Fama and French classification that
financial distress than bankruptcy (Grice & transfers the SIC classification codes into 49 industry
Ingram, 2001). groups.
Z = 1.2 (working capital divided by total assets) + 1.4 We notice that our sample is formed mainly by firms
(retained earnings divided by total assets) + 3.3 (earnings from the following industries: Pharmaceutical Products,
before interest and taxes divided by total assets) + 0.6 Petroleum and Natural Gas and Computer Software.
(market value of equity divided by total liabilities) + 0.999 The Panel A of Table 2 represents the mean, min and
(sales divided by total assets). max of the continuous variables. We note that religiosity
We identify a new variable namely Z score_1968 to varies between 0, 28 and 0, 77 with a low SD 0, 11.
measure the financial distress (Bhaskar et al., 2017). Panel B represents the distribution of our sample by
Z score_1968 = 0 if Z > 3 Safe zone. financial health. We notice that 48% of firms are healthy
Z score_1968 = 1 if 1.81 <Z< 2.99 Grey zone. and 32% of firms are in financial distress.
Z score_1968 = 2 if Z < 1.81 Distress zone. Table 3 notes the frequency of audit opinion. Our
A higher Z score_1968 indicates a higher probability sample is mainly composed of firms with unqualified
of distress. opinion 70, 37%. This opinion means that financial state-
We control for some firm's characteristics. First, an ments comply with U.S. GAAP and these statements rep-
unsatisfactory audit opinion is a negative signal. Hudaib resent fairly the financial situation of the firm.
and Cooke (2005) find that financially distressed firms
are more likely to receive a qualified audit report from
large audit firms. 6.2 | Univariate tests and correlations
We also control for the company's debt position
(Opler & Titman, 1994), firm size (Rajan & In order to study the binary associations between indepen-
Zingales, 1995) and we include the year and industry dent variables, we run a correlation matrix using a correla-
dummies to control for fixed time and industry effects. tion test. The correlation coefficients between independent
We include a set of county-level demographic charac- variables enable to explore the presence of a severe prob-
teristics from the U.S. Census Bureau in our analysis as lem of multicollinearity. This problem is detected when
control variables (Hilary & Hui, 2009). The variables are the coefficient is higher than 0.8 (Franke, 2010).
the total population of the county; education, measured Table 4 shows that correlation coefficients between
as the proportion of county population above age 25 that independent variables are lower than 0.8, except Log_-
has completed a bachelor's degree or higher; income, percapita income and education (0.799). For this reason,
which is the per capita personal income; male–female we use the variance inflation factor (VIF) method to
ratio, measured as the male population to the female detect a serious problem of multicollinearity. The results
population; Married, measured as percentage of married show that there is no severe problem (Table 5).
people in the county.
We expect a positive effect of total population,
income and education on financial distress. While we 7 | MAIN MULTIVARIATE
predict a positive impact of male–female ratio and a ANALYSIS
negative impact of a percentage of married people on
financial distress because of the risk aversion. These We use panel data to study the relation between Z-score
demographic characteristics are from a census done in and religiosity over the years and across firms. We will
ten-year intervals. We linearly interpolate these vari- select our estimator based on the results of the homogene-
ables for the intermediate years. The Appendix spec- ity test. With clustered standard errors, we obtain a vari-
ifies all the variables in our regressions with data ance estimate that is robust to potential heteroscedasticity
sources. and autocorrelation problems.
GHARBI ET AL. 7

TABLE 1 Distribution of firms by industry TABLE 1 (Continued)

Industry Freq. Percentage Cum. Industry Freq. Percentage Cum.

Agriculture 187 0.24 0.24 Retail 2,226 2.84 95.26
Food products 1,129 1.44 1.68 Restaurants, hotels, motels 1,546 1.97 97.23
Candy and soda 246 0.31 1.99 Almost nothing 2,168 2.77 100.00
Beer and liquor 309 0.39 2.39 78,317 100.00
Tobacco products 122 0.16 2.54
Recreation 846 1.08 3.63
7.1 | Regression results
Entertainment 1,225 1.56 5.19
Printing and publishing 802 1.02 6.21 Given that the homogeneity test results lead to confirm
Consumer goods 1,478 1.89 8.10 the presence of individual effects (Table 6), we estimate
Apparel 1,139 1.45 9.55 our coefficients using OLS regression with clustered stan-
Healthcare 1,857 2.37 11.93 dard errors at the firm level. All regressions control for a
Medical equipment 2,640 3.37 15.30 year and Fama French 49 industry fixed effect.
The univariate analysis suggests that the degree of a
Pharmaceutical products 5,699 7.28 22.57
county's religiosity is negatively related to the degree of
Chemicals 1,696 2.17 24.74
financial distress.
Rubber and plastic products 1,042 1.33 26.07 We now examine whether these relations persist after
Textiles 617 0.79 26.86 controlling for other potential determinants of financial
Construction materials 2,570 3.28 30.14 distress and a set of county-level demographic
Construction 803 1.03 31.16 characteristics.
We estimate the following regression:
Steel works, etc. 1,236 1.58 32.74
ZSCORE_1968 it = β0 + β1 RELIGIOSITYit + β2
Fabricated products 429 0.55 33.29
AUDITOPNit + β3 LEVERAGEit + β4 SIZEit + β5 popu-
Machinery 3,209 4.10 37.39 lation it + β6 education it + β7 married it + β8 per-
Electrical equipment 1,312 1.68 39.06 capita_income it + β9male_female it + +β10 Yrit + β11
Automobiles and trucks 1,324 1.69 40.75 Indit + ε (see the Appendix A for the variables
Aircraft 534 0.68 41.44 definitions).
Shipbuilding, railroad equipment 91 0.12 41.55 Panel A of Table 7 presents the outputs of regressions
of financial distress for the sample. In column 3, religiosity
Defence 190 0.24 41.79
has a negative coefficient in explaining financial distress
Precious metals 512 0.65 42.45
statistically significant at the 1% level. Consistent with our
Non-metallic and industrial 554 0.71 43.16 hypothesis, this result suggests that firms in more religious
metal mining
areas exhibit lower financial problems than the others
Coal 184 0.23 43.39 firms after controlling for a firm-specific and county- spe-
Petroleum and natural gas 5,240 6.69 50.08 cific variables. In economic terms, the estimated coeffi-
Utilities 3,064 3.91 53.99 cient of −0.155 on religiosity suggests that one SD increase
Communication 3,264 4.17 58.16 in the fraction of religious members leads to a decrease of
about 0.018 (= − 0.155*0.117) in financial trouble.
Personal services 732 0.93 59.10
Our result confirms the works of two lines of research.
Business services 4,422 5.65 64.74
The first one suggests that religious beliefs have an impact
Computers 2,191 2.80 67.54 on corporate decisions and thus on financial output. In
Computer software 5,923 7.56 75.10 fact, religious firms are less likely to have tax avoidance, to
Electronic equipment 4,182 5.34 80.44 engage on fraud, to manage their earning (Cai, Kim, Li, &
Measuring and control equipment 1,819 2.32 82.76 Pan, 2019). For this reason, they are less likely to have
Business supplies 1,423 1.82 84.58
financial distress. If we begin with the idea that the reli-
gious culture prevents firms from corporate misbehaviour
Shipping containers 400 0.51 85.09
due to the ethics (Benmelech & Frydman, 2015;
Transportation 2,900 3.70 88.80
Biggerstaff, Cicero, & Puckett, 2015; Davidson, Dey, &
Wholesale 2,835 3.62 92.42 Smith, 2015), and encourages the aversion risk behaviour
(Continues) so it minimizes many costs related to agency costs,
8 GHARBI ET AL.

TABLE 2 Descriptive statistics

Panel A
Variable Obs Mean SD Min Max
Religiosity 78,317 .523 .117 .284 .777
Leverage 78,317 .257 .249 0 1.335
Size 78,317 4.811 2.229 .392 10.25
Log_population 78,317 13.608 1.158 7.321 16.1
Log_percapita income 78,317 9.873 .515 8.113 11.003
Male_female 78,317 .956 .04 .852 1.261
Education 78,317 28.769 10.254 4.031 68.8
Married 78,317 .766 .066 .443 .924
Panel B
Z_score_1968 Freq. Percentage
0 37,673 48.10
1 15,356 19.61
2 25,288 32.29
Total 78,317 100.00

TABLE 3 Frequency of audit opinion (significant at the 1% level). However, when we introduce
Opinion Freq. Percentage Cum.
religiosity, the coefficient of “religiosity*crisis_2008” is
negative and statistically significant. Consistent with our
0 216 0.28 0.28
hypothesis, this result suggests that firms in more reli-
1 55,110 70.37 70.64 gious areas exhibit lower financial distress in the finan-
2 2,373 3.03 73.67 cial crisis of 2008.
3 129 0.16 73.84 In others terms, religiosity constituted a solution to
4 20,486 26.16 100.00 curb the financial problems during the crisis.
5 3 0.00 100.00 This result confirms the idea that financial system
in the United States has a lack of ethics. Nelson (2017)
Total 78,317 100.00
notes that the financial crisis had a religious origin. In
the same way, Lewis, Kay, Kelso, and Larson (2010)
consider that the lack of morals caused the financial
lawsuits costs. In the same order of idea, Leventis, crisis. So, religious firms are less likely to suffer from
Dedoulis, and Abdelsalam (2018) find that religious firms financial problems in 2008 than the others firms
are less likely to have agency costs. For this reason, the because they have more ethics and less risk taking.
audit pricing is negatively related to religiosity. Adhikari and Agrawal (2016) note that banks in more
The second line of research is related to the good rep- religious counties are less likely to take risk in the
utation of religious firms. In fact, trust constitutes a guar- period of crisis than the others banks and conse-
antee for many stakeholders. For example, Kim quently, they are less likely to have troubles.
et al. (2018) consider that religiosity favourize a corporate
social responsibility, He and Hu (2016) suggest that reli-
gious firms have a favourite loan conditions, and El 7.2 | Robustness
Ghoul, Guedhami, Ni, Pittman, and Saadi (2012) find
that religious firms benefit from a lower cost of equity We next conduct a set of robustness test to verify if our
than the no religious firms. All these arguments argue finding persists. For this reason, we utilize another mea-
that the good image of religious firms offer them a great sure of financial distress: the revised Altman Z" score. In
work conditions and then a good performance. fact, the original Altman Z-score was used for manufactur-
The objective of Panel B is to see the effect of religios- ing firms. For this reason, in 1995, Altman was introduced
ity in the period of crisis. For this reason, we present this a revised score to fit different sectors.
variable “religiosity*crisis_2008.” We note that the vari- The difference between Z-score and Z" is the elimination
able “year2008” is positively related to financial distress of the ratio sales/total assets. In fact, this ratio varies in the
GHARBI ET AL. 9

TABLE 4 Correlation matrix

Pairwise correlations

Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
(1) z_score_1968 1.000
(2) religiosity −0.005 1.000
0.185
(3) leverage 0.504 0.032 1.000
0.000 0.000
(4) size 0.010 0.043 0.081 1.000
0.006 0.000 0.000
(5) population 0.046 0.034 −0.033 0.042 1.000
0.000 0.000 0.000 0.000
(6) Percapita income 0.120 0.092 −0.049 0.083 0.188 1.000
0.000 0.000 0.000 0.000 0.000
(7) Male_female 0.046 −0.437 −0.044 −0.062 0.039 0.175 1.000
0.000 0.000 0.000 0.000 0.000 0.000
(8) married −0.101 −0.103 −0.031 −0.140 −0.442 −0.334 0.281 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
(9) education 0.066 0.068 −0.075 0.053 0.191 0.799 0.082 −0.184 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(10) Auditor_opinion 0.216 0.012 0.134 0.085 0.046 0.292 0.069 −0.156 0.171 1.000
0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000

TABLE 5 Variance inflation factor TABLE 6 Homogeneity test

VIF 1/VIF FISHER Sig

Percapita income 3.808 .263 8,42 0.0000

Education 3.063 .326
Married 1.795 .557
Male_female 1.672 .598 where X1 is the working capital/total assets, X2 is the
Population 1.348 .742 retained earnings/total assets, X3 is the EBIT/total assets,
Religiosity 1.322 .756 and X4 is the book value equity/total liabilities.
Auditor opinion 1.138 .879 We introduce another variable namely Z" score_1995
Leverage 1.039 .963 to measure financial distress.
Z"score_1995 = 0 if Z" > 2.60 Safe zone.
Size 1.032 .969
Z“score_1995 =1 if 1.10 <Z"< 2.60 Grey zone.
Mean_VIF 1.802 .969
Z“score_1995= 2 if Z" < 1.10 Distress zone.
In Table 8 (Panel A), we can see that the negative
various sectors. For example, for high-tech firms, the ratio relationship between religiosity and financial distress is
sales/total assets are higher than manufacturing firms still robust after using the revised Altman Z “score. The
because total assets are very lower than sales. For this reason, negative coefficient is significant at a 1% level. The effect
we use, in this section, the Z" score to fit more all sectors. of the crisis on financial distress is positive and very sub-
The formula of Z" is as follows (Altman, Hartzell, & stantial (Panel B). However, this coefficient becomes neg-
Peck, 1998): ative when we introduce the impact of religiosity.
Despite using the revised Z “score, the robustness test
Z00 = 6:56X1 + 3:26X2 + 6:72X3 + 1:05X4 confirms our results about the impact of religiosity on the
10 GHARBI ET AL.

TABLE 7 Regressions results

Panel A impact of religiosity on financial distress

z_score_1968 Predict sign Coef. SE t-value p-value Sig.

Religiosity _ −0.155 0.058 −2.69 .007 ***
Auditor_opinion + 0.076 0.003 22.86 .000 ***
Leverage + 1.628 0.022 72.50 .000 ***
Size _ −0.049 0.003 −14.54 .000 ***
Log_population 0.025 0.006 3.89 .000 ***
Log_percapita_income 0.043 0.059 0.72 .471
Education 0.002 0.002 1.59 .112
Married −0.311 0.147 −2.12 .034 **
Male_female 0.150 0.211 0.71 .476
Year effect yes Yes
Industry effect yes Yes
Cluster cusip yes Yes
R-squared 0.383 Number of obs 78,317.000
F-test 196.764 Prob > F 0.000
Panel B impact of religiosity on financial distress in the crisis period
z_score_1968 Coef. St.err t-value p-value Sig.
Religiosity −0.142 0.058 −2.46 0.014 **
Auditor opinion 0.076 0.003 22.85 0.000 ***
Leverage 1.628 0.022 72.49 0.000 ***
Size −0.049 0.003 −14.54 0.000 ***
Log_population 0.025 0.006 3.90 0.000 ***
Log_percapita_income 0.042 0.059 0.71 0.480
Education 0.002 0.002 1.60 0.110
Married −0.312 0.147 −2.12 0.034 **
Crisis 2008 0.280 0.084 3.33 0.001 ***
Religiosity*Crisis 2008 −0.424 0.160 −2.65 0.008 ***
Year effect yes Yes
Industry effect yes Yes
Cluster cusip yes Yes

R-squared 0.383 Number of obs 78,317.000

F-test 192.445 Prob > F 0.000
***
p < 0.01,
**
p < 0.05,
*
p < 0.1.

financial stability of U.S. firms and especially in the argue that the religion has an impact on the growth rate in
period of crisis. United States and Guiso et al. (2008) explain why the reli-
gion similarity between two countries facilities the trade vol-
ume. Another line of research was developed to explain
8 | C ON C L U S I ON the impact of religiosity on the financial and investment
decisions due to ethics and risk aversion. Moreover, some
Religiosity seems to be very important in the economic researches find that religiosity has an impact on stock
development of the countries. Rupasingha and Chilton (2009) price crash risk (Callen & Fang, 2015). In other word, it
GHARBI ET AL. 11

TABLE 8 Robustness tests

Panel A: Impact of religiosity on financial distress using Z"_score_1995

Z"_score_1995 Coef. SE t-value p-value Sig.

Religiosity −0.172 0.060 −2.86 .004 ***
Auditor_opinion 0.085 0.003 25.86 .000 ***
Leverage 1.461 0.022 66.57 .000 ***
Size −0.105 0.003 −31.04 .000 ***
Log_population 0.028 0.006 4.37 .000 ***
Log_percapita_income 0.059 0.061 0.97 .332
Education 0.003 0.002 2.21 .027 **
Married −0.353 0.145 −2.44 .015 **
Male_female 0.161 0.217 0.74 .458
Year effect yes Yes
Industry effect yes Yes
Cluster cusip yes Yes
R-squared 0.376 Number of obs 78,316.000
F-test 178.945 Prob > F 0.000
Panel B: Impact of religiosity on financial distress during the crisis period using
Z"_score_1995

Z"_score_1995 Coef. SE t-value p-value Sig.

Religiosity −0.159 0.060 −2.63 .008 ***
Auditor_opinion 0.085 0.003 25.84 .000 ***
Leverage 1.462 0.022 66.57 .000 ***
Size −0.105 0.003 −31.03 .000 ***
Log_population 0.028 0.006 4.38 .000 ***
Log_percapita_income 0.058 0.061 0.96 .338
Education 0.003 0.002 2.22 .026 **
Married −0.354 0.145 −2.44 .015 **
Male_female 0.168 0.217 0.77 .440
Crisis 2008 0.194 0.077 2.53 .011 **
Religiosity*Crisis 2008 −0.419 0.147 −2.85 .004 ***
Year effect yes Yes
Industry effect yes Yes
Cluster cusip yes Yes
R-squared 0.376 Number of obs 78,316.000
F-test 175.068 Prob > F 0.000

influences the firm's continuity. So, it is interesting to see after controlling a set of county-level characteristics and
if the financial troubles are also related to the religiosity of firm-level variables. Moreover, these results persist after
the firm. Many papers discussed the causes of financial using the modified Z" instead of the original Altman Z-
trouble like leverage, tax avoidance, accounting manipula- score as a measure of financial distress.
tion. However, these causes are driven by the human Our study implies the importance of religiosity. In
behaviour. This paper contributes to the literature by dis- fact, it determines the financial stability of the firm. So,
covering the effect of religiosity on the financial distress of without ethics and with a lot of risk, firms can disappear.
U.S. firms. Using 78,317 observations, we find that the reli- Our finding extends the previous papers. First, it com-
giosity influences negatively the financial difficulty. This pletes the line of research about the role played by a
result becomes stronger in the crisis period, especially in behavioural factor such as religiosity which composed by
the crisis of 2008. This negative relation remains robust values and beliefs. Second, it explains another cause of
12 GHARBI ET AL.

financial distress during crisis. For this reason, these results Center series on financial markets and institutions (Vol. 2). Bos-
should be of interest of regulators because they can explain ton, MA: Springer.
the effect of human behaviour during the crisis and why Barro, R. J., & McCleary, R. M. (2003). Religion and economic growth
across countries. American Sociological Review, 68(5), 760–781.
some firms are more sensitive than others. In the same
Benmelech, E., & Frydman, C. (2015). Military CEOs. Journal of
order of idea, many reasons were presented to explain the Financial Economics, 117(1), 43–59.
financial crisis such as economics financials and politics. Bertrand, M., & Schoar, A. (2003). Managing with style: The effects
However, all these arguments have two ultimate origins: of managers on firm policies. Quarterly Journal of Economics,
the lack of ethics and the excessive risk taking. Conse- 118, 1169–1208.
quently, it is not enough to enact laws to prevent the crisis Bhaskar, L. S., Krishnan, G. V., & YU, W. (2017). Debt covenant
but the rules must be made with morality and ethics. violations, firm financial distress, and auditor actions. Contem-
porary Accounting Research, 34, 186–215.
Our study suffers from a major limitation related to
Biggerstaff, L., Cicero, D. C., & Puckett, A. (2015). Suspect CEOs,
the data availability. The period of the study is from 1995
unethical culture, and corporate misbehavior. Journal of Finan-
to 2010 due to the data related to the religiosity. In fact, cial Economics, 117(1), 98–121.
ARDA published every 10 years the data. For this reason, Boone, J. P., Khurana, I. K., & Raman, K. K. (2013). Religiosity and tax
we are not able to study the effect of religiosity after crisis avoidance. Journal of the American Taxation Association, 35, 53–84.
in long term and see if its impact is still working. More- Cai, Y., Kim, Y., Li, S., & Pan, C. (2019). Tone at the top: CEOs' reli-
over, testing if religiosity can prevents the bankruptcy gious beliefs and earnings management. Journal of Banking &
especially after crisis will be very useful, which we can Finance, 106(C), 195–213.
Callen, J. L., & Fang, X. (2015). Religion and stock price crash risk.
suggest for future study.
Journal of Financial and Quantitative Analysis, 50, 169–195.
Chen, J., De Cesari, A., Hill, P., & Ozkan, N. (2018). Initial compen-
ORCID sation contracts for new executives and financial distress risk:
Khaled Hussainey https://orcid.org/0000-0003-3641- An empirical investigation of UKfirms. Journal of Corporate
1031 Finance, 48, 292–313.
Cialdini, R. (1993). Influence: The psychology of persuasion.
E N D N O T ES New York, NY: Quill.
i Cialdini, R. B., & Goldstein, N. J. (2004). Social influence: Compliance
Using the website http://www.zipcodestogo.com/county-zip-code-
and conformity. Annual Review of Psychology, 55(1), 591–621.
list.htm
Davidson, R. H., Dey, A., & Smith, A. J. (2015). (2015). Executives'
1
https://www.economy.com/dismal/analysis/104331 “off-the-job” behavior, corporate culture, and financial
2
https://news.gallup.com/poll/114022/state-states-importance- reporting risk. Journal of Financial Economics, 117(1), 5–28.
religion.aspx Donker, H., Ng, A., & Shao, P. (2018). Borrower distress and the
efficiency of relationship banking. Journal of Banking &
Finance, 000, 1–17.
DATA AVAILABILITY STATEMENT Du, X. (2012). Does religion matter to owner-manager agency costs?
Evidence from China. Journal of Business Ethics, 118(2), 319–347.
We have mentioned to sources of data in our paper.
Duarte, J., Siegel, S., & Young, L. (2012). Trust and credit: The role
of appearance in peerto-peer lending. Review of Financial Stud-
ORCID ies, 25, 2454–2483.
Khaled Hussainey https://orcid.org/0000-0003-3641- Dyreng, S. D., Mayew, W. J., & Williams, C. D. (2012). Religious
1031 social norms and corporate financial reporting. Journal of Busi-
ness Finance & Accounting, 39(7–8), 845–875.
R EF E RE N C E S El Ghoul, S., Guedhami, O., Ni, Y., Pittman, J., & Saadi, S. (2012).
Does religion matter to equity pricing? Journal of Business
Adhikari, B. K., & Agrawal, A. (2016). Does local religiosity matter
Ethics, 111(4), 491–518.
for Bank risk-taking? Journal of Corporate Finance, 38,
Fairfield, P. M., Whisenant, J. S., & Yohn, T. L. (2003). Accrued
272–293.
earnings and growth: Implications for future profitability and
Ahmad, M. (1973). Religiosity as a function of rigidity and anxiety.
market mispricing. The Accounting Review, 78(1), 353–371.
Indian Journal of Experimental Psychology, 7, 49–50.
Fama, E. F., & French, K. R. (1992). The cross-section of expected
Ahmed, P. K., Loh, A. Y. E., & Zairi, M. (1999). Cultures for contin-
stock returns. The Journal of Finance, 47, 427–465.
uous improvement and learning. Total Quality Management, 10
Franke, G. R. (2010). Multicollinearity. Wiley International encyclo-
(4–5), 426–434.
pedia of marketing, New Jersey, United States: . John Wiley &
Altman, E. I. (1968). Financial ratios, discriminant analysis and the
Sons. https://doi.org/10.1002/9781444316568.wiem02066.
prediction of corporate bankruptcy. The Journal of Finance, 23,
Gao, L., Wang, Y., & Zhao, Y. (2017). Does local religiosity affect
589–609.
organizational risk-taking? Evidence from the hedge fund
Altman, E., Hartzell, J., & Peck, M. (1998). Emerging markets cor-
industry. Journal of Corporate Finance, 47, 1–22.
porate bonds: A scoring system. In R. M. Levich (Ed.), Emerg-
Gibson, D. (2017). Doing good and doing well: Faith-based
ing market capital flows. The new York University Salomon
investing converts the skeptics. Religion news service, May 31.
GHARBI ET AL. 13

Greif, A. (1994). Cultural beliefs and the organization of society: A Landes, D. (2000). Culture makes almost all the difference. In L. E.
historical and theoretical reflection on collectivist and individu- Harrison & S. P. Huntington (Eds.), Culture matters. How values
alist societies. Journal of Political Economy, 102, 912–950. shape human progress, (2–13). New York, NY.Basic Books.
Grice, J. S., & Ingram, R. W. (2001). Tests of the generalizability of Leng, J., Trzeciakiewicz, A., & Ozkan, A. (2018). CEO Over-
Altman's bankruptcy prediction model. Journal of Business confidence and the Probability of Corporate Failure: Evidence
Research, 54, 53–61. from the UK. Working paper. (February 28, 2018). Available at
Grullon, G., Kanatas, G., & Weston, J. (2010). Religion and corpo- SSRN: https://ssrn.com/abstract=3184199
rate (Mis) behavior. Working paper, USA: . Rice University. Leventis, S., Dedoulis, E., & Abdelsalam, O. (2018). The impact of reli-
Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_ giosity on audit pricing. Journal of Business Ethics, 148(1), 53–78.
id=1472118 Lewis, V., Kay, K. D., Kelso, C., & Larson, J. (2010). Was the 2008
Guiso, L., Sapienza, P., & Zingales, L. (2003). People's opium? Reli- financial crisis caused by a lack of corporate ethics? Global
gion and economic attitudes. Journal of Monetary Economics, Journal of Business Research, 4(2), 77–84.
50, 225–282. Li, X., Wang, S. S., & Wang, X. (2017). Trust and stock Price crash risk:
Guiso, L., Sapienza, P., & Zingales, L. (2006). Does culture affect eco- Evidence from China. Journal of Banking and Finance, 76, 74–91.
nomic outcomes? Journal of Economic Perspectives, 20(2), 23–48. Lo, A.W. (2008). Hedge funds, systemic risk, and the financial crisis
Guiso, L., Sapienza, P., & Zingales, L. (2008). Trusting the stock of 2007–2008. Testimony to the U.S. House of Representatives,
market. Journal of Finance, 63(6), 2557–2600. Committee on Oversight and Government Reform, Hearing on
Guiso, L., Sapienza, P., & Zingales, L. (2009). Culture biases in economic Hedge Funds, November 13.
exchange? Quarterly Journal of Economics, 124(3), 1095–1131. Longenecker, J. G., McKinney, J. A., & Moore, C. W. (2004). Reli-
Guiso, L., Sapienza, P., & Zingales, L. (2015). The value of corporate gious intensity, evangelical Christianity, and business ethics:
culture. Journal of Financial Economics, 117(1), 60–76. An empirical study. Journal of Business Ethics, 55(4), 373–386.
Hanlon, M., & Slemrod, J. (2009). What does tax aggressiveness sig- Malmendier, U., Tate, G., & Yan, J. (2011). Overconfidence and
nal? Evidence from stock price reactions to news about tax shel- early-life experiences: The effect of managerial traits on corpo-
ter involvement. Journal of Public Economics, 93(1), 126–141. rate financial policies. Journal of Finance, 66, 1687–1733.
He, W., & Hu, M. R. (2016). Religion and bank loan terms. Journal McAfee, R.P., (2004). The real lesson of Enron's implosion: Market
of Banking and Finance, 64, 205–215. makers are in the trust business. The Economists' Voice, 1(2),
Hilary, G., & Hui, K. W. (2009). Does religion matter in corporate Article4.1–10.
decision making in America? Journal of Financial Economics, McGuire, S. T., Omer, T. C., & Sharp, N. Y. (2012). The impact of
93, 455–473. religion on financial reporting irregularities. The Accounting
Hirshleifer, D. (2014). Behavioral finance. Working paper, Ivrine Review, 87(2), 645–673.
California, United States: . University of California, Irvine – Merchant, K. A., & Rockness, J. (1994). The ethics of managing
Paul Merage School of Business; NBER. Available at SSRN: earnings: An empirical investigation. Journal of Accounting and
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2480892 Public Policy, 13(1), 79–94.
Hudaib, M., & Cooke, T. E. (2005). The impact of managing director Nelson, R. H. (2017). The financial crisis as a religious crisis. Jour-
changes and financial distress on audit qualification and audi- nal of International Business and Law, 17(1).45–86. https://
tor switching. Journal of Business Finance and Accounting, 32 scholarlycommons.law.hofstra.edu/jibl/vol17/iss1/5.
(9–10), 1703–1739. North, D. C. (1990). Institutions, institutional change and economic
Jiang, F., John K., Li C. W., Qian Y. (2018). Earthly Reward to the performance. Cambridge: Cambridge University Press.
Religious: Religiosity and the Costs of Public and Private Debt. Omer, T. C., Sharp, N. Y., & Wang, D. (2016). The impact of religion
Journal of Financial and Quantitative Analysis, 53, (5), 2131– on the going concern reporting decisions of local audit practice
2160. http://dx.doi.org/10.1017/s002210901800039x. offices. Journal of Business Ethics, 149(4), 811–831.
Kanagaretnam, K., Lim, C. Y., & Lobo, G. J. (2014). Influence of Opler, T. C., & Titman, S. (1994). Financial distress and corporate
National Culture on accounting conservatism and risk-taking in performance. Journal of Finance, 49, 1015–1040.
the banking industry. The Accounting Review, 89(3), 1115–1149. Pirinsky, C., & Wang, Q. (2006). Does corporate headquarters loca-
Kennedy, E. J., & Lawton, L. (1998). Religiousness and business tion matter for stock returns? Journal of Finance, 61, 1991–2015.
ethics. Journal of Business Ethics, 17(2), 163–175. Rajan, R., & Zingales, L. (1995). What do we know about capital
Kim, H., Yadav, P., Han, S. H., & Yang, L. (2018). Local Religiosity structure? Some evidence from international data. The Journal
and Corporate Social Responsibility. Working paper. (January of Finance, 50, 1421–1460.
14, 2018). Available at SSRN: https://ssrn.com/abstract=3101612. Richardson, G., Lanis, R., & Taylor, G. (2015). Financial distress,
Kumar, A., Page, J. K., & Spalt, O. G. (2011). Religious beliefs, gam- outside directors and CorporateTax aggressiveness spanning
bling attitudes, and financial market outcomes. Journal of the global financial crisis: An empirical analysis. Journal of
Financial Economics, 102(3), 671–708. Banking and Finance, 52, 112–129.
Lagace, M. (2001). Can religion and business learn from each other?, Rijsenbilt, A., & Commandeur, H. (2013). Narcissus enters the
Boston, Massachusetts, United States: Harvard Business School courtroom: CEO narcissism and fraud. Journal of Business
Working Knowledge. (November 12). Available at: http:// Ethics, 117(2), 413–429.
hbswk.hbs.edu/item/3511.html. Rokeach, M. (1968). Beliefs, attitudes, and values. San Francisco:
Lal, D. (1999). Unintended consequences: The impact of factor Jossey-Bass Inc.
endowments, culture, and politics on long-run economic perfor- Stulz, R. M., & Williamson, R. (2003). Culture, openness, and
mance. Cambridge, MA: MIT Press. finance. Journal of Financial Economics, 70, 313–349.
14 GHARBI ET AL.

Rupasingha, A., & Chilton, J.B., (2009). Religious adherence and Wallis, J. (2010). A religious response to the financial crisis. Wash-
county economic growth in the US. Journal of Economic Behav- ington post, January 3.
ior & Organization, vol. 72(1), pages 438–450, October.
Speiser, M. (2015). This map shows the US really has 11 separate
'nations' with entirely different cultures. Business Insider, July 27.
Sonnad, N. (2018). The most memorable phrase of the financial cri-
sis taught us the wrong lesson. Quartz, September 13.
How to cite this article: Gharbi I, Hamed-
Sunstein, C. R. (1996). Social norms and social rules. Columbia Law
Sidhom M, Hussainey K, Ganouati J. Religiosity
Review, 96(4), 903–968.
Tergesen, A. (2017). Financial Advisers Put Faith in Religion-Based and financial distress in U.S. firms. Int J Fin Econ.
Investing. The Wall Street Journal, March, Updated April 5. 2020;1–14. https://doi.org/10.1002/ijfe.1994
Tonge, A., Greer, L., & Lawton, A. (2003). The Enron story: You
can fool some of the people some of the time. Business Ethics: A
European Review, 12(1), 4–22.
Weaver, G. R., & Agle, B. R. (2002). Religiosity and ethical behavior
in organizations: A symbolic interactionist perspective. Acad- APP E ND IX A : Variable definitions
emy of Management Review, 27(1), 77–97.

Measures Sources
Dependent variable ZSCORE_1968: COMPUSTAT North America
Z score_1968 = 0 if Z > 3, Z
score_1968 = 1 if 1.81 <Z< 2.99 and
Z score_1968 = 2 if Z < 1.81,
Independent variable
Religiosity Ratio of religious members to the American religion data archive (ARDA)
population of the county where the website
firm is headquartered
Control variables
County-level demographic Population: Natural logarithm of Total U.S. Census Bureau
characteristics population of the county.
Education: Proportion of county
population above age 25 that has
completed a bachelor's degree or
higher.
Income: Natural logarithm of the per
capita personal income.
Male–female ratio: Measured as the
male population to the female
population.
Married: The percentage of married
people in the county
Leverage Ratio of total debt/total assets COMPUSTAT North America
Size Natural logarithm of total assets COMPUSTAT North America
Audit opinion 0 financial statements are unaudited COMPUSTAT North America
1 unqualified opinion.
2 qualified opinion.
3 no opinion.
4 unqualified opinion with explanatory
language.
5 adverse opinion. Auditor has
expressed an adverse opinion
regarding the financial statements of
the company
Yr Year dummy variables COMPUSTAT North America
Ind Industry dummy variables