Does lesbian and gay friendliness pay off? A new look at
LGBT policies and firm performance
Veda Fatmya, John Kihna, Jukka Sihvonenb , Sami Va¨ha¨maaa
aSchool of Accounting and Finance, University of Vaasa, Vaasa,
bSchool of Business, Aalto University, Espoo, Finland
Abstract
This paper examines the association between LGBT-friendly corporate policies
and firm performance. Using data on US firms from 2003 to 2016, we
document that LGBT friendliness is positively associated with firm perfor-
mance. Specifically, we find strong evidence that more LGBT-friendly firms
have higher profitability and higher stock market valuations. Our results
further demonstrate that the positive effect of progressive LGBT policies on
firm performance is more pronounced for firms located in more liberal states.
Overall, our empirical findings provide support for the view that socially
progressive corporate policies and diversity management may create value for
the firm.
Key words: LGBT policies; Firm performance; Sexual minorities; Employee
policies
JEL classification: D22, G30, G31, G39, J15, J70, J83, M12, M50
doi: 10.1111/acfi.12787
We would like to thank Audur Arna Arnardottir, Kirsten Cook, Jonathan Fluharty-
Jaidee, Greg Filbeck, Markku Kaustia, Christos Pantzalis, Jung Chul Park, Pontus
Troberg, and conference and seminar participants at the 2019 Annual Meeting of the
Southern Finance Association, the 2018 Annual Meeting of the Eastern Finance
Association, the 2017 Diversity Section Meeting of the American Accounting Associ-
ation, the 2017 Annual Workshop of the Nordic Corporate Governance Network, the
Hanken School of Economics Workshop on Corporate Governance, the 2017 Finnish
Graduate School of Finance Summer Workshop, the University of South Florida, and
the University of Vaasa for valuable comments and suggestions. This work was
supported by the OP Group Research Foundation, the Marcus Wallenberg Foundation,
the Finnish Savings Banks Research Foundation, and the Foundation for Economic
Education. Part of this paper was written while S. Va¨ha¨maa was visiting Florida
Atlantic University and the University of Deusto.
Please address correspondence to Sami Va¨ha¨maa via email: sami@univaasa.fi
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia, Ltd on
behalf of Accounting and Finance Association of Australia and New Zealand
This is an open access article under the terms of the Creative Commons Attribution-NonCommerc
ial-NoDerivs License, which permits use and distribution in any medium, provided the original
work is properly cited, the use is non-commercial and no modifications or adaptations are made.
Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1. Introduction
Corporate social advocacy can be a tricky business. While taking a public
stand on potentially sensitive social or political issues may lead to positive
outcomes and competitive advantages, the repercussions of social advocacy can
also be detrimental if the stance taken is not aligned with the preferences and
values of the firm’s key stakeholders. One particularly visible and divisive form
of corporate social advocacy in the US has been firms’ engagement in the
public socio-political debate related to sexual minorities. Over the past decade,
many prominent large firms have been among the most visible proponents of
lesbian, gay, bisexual and transgender (LGBT) rights and anti-discrimination
policies despite the risk of potentially alienating some of their employees,
customers and other stakeholders who may share different social values.1 Given
that LGBT advocacy often has no intrinsic relation to firms’ core business
operations, why do firms engage in a controversial debate over sexual
minorities instead of remaining value-neutral? The natural question then is
whether LGBT friendliness pays off for the firms. In this paper, we address this
question by empirically examining the relationship between LGBT-friendly
corporate policies and firm performance.
There are various reasons why LGBT friendliness may influence firm
performance. In general, corporate social advocacy such as the adoption of
LGBT-supportive policies or taking a stand on same-sex marriage can be
broadly considered as an element of corporate social responsibility (see, e.g.,
Snider et al., 2003; Weinzimmer and Esken, 2016; Wettstein and Baur, 2016;
Shan et al., 2017). Consequently, the theoretical arguments for a link between
corporate social responsibility (CSR) and firm performance are largely
applicable also in the context of LGBT advocacy. While the classical
shareholder-oriented view of Friedman (1962, 1970) posits that the only social
responsibility of a firm is to increase its profits, implying that CSR is potentially
value-destroying expenditure of corporate resources, the stakeholder theory
established by Freeman (1984) argues that engagement in social activities
creates shareholder value by forging relationships with the firm’s key
stakeholders.2 In the vast body of CSR literature, the stakeholder theory has
1Over recent years, prominent firms such as Apple, Coca-Cola, Goldman Sachs, Google,
Hewlett-Packard, Intel, KPMG, PwC, Starbucks, Target and Walt Disney have engaged
in the public discussion and have actively supported sexual minorities (see, e.g., Aspan,
2020).
2Consistent with Friedman’s (1970) shareholder-oriented view, Henderson (2001),
Jensen (2002) and Sundaram and Inkpen (2004) have criticised socially responsible
behaviour as an unnecessary cost and potentially value-destroying investment.
Freeman’s (1984) stakeholder view is comprehensively discussed and further elaborated,
e.g., in Donaldson and Preston (1995), Jones (1995), Agle et al. (2008) and Laplume
et al. (2008).
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
214 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
become the central paradigm for rationalising why social responsibility may
pay off by enhancing firm reputation, customer relationships, accumulation of
human capital and access to resources and external financing (see, e.g.,
Waddock and Graves, 1997; Barnett, 2007; Artiach et al., 2010; Surroca et al.,
2010; Faleye and Trahan, 2011; Barnett and Salomon, 2012; Flammer and
Luo, 2017; Buchanan et al., 2018).3 Thus, if LGBT friendliness is not
conflicting with stakeholders’ expectations and values, the stakeholder theory
predicts a positive relationship between LGBT-friendly policies and firm
performance.
Loosely parallel with the stakeholder view, the human resource management
(HRM) theories regarding employee satisfaction and diversity management
provide an alternative motivation for hypothesising a positive linkage between
LGBT-friendly policies and firm performance. These theories recognise
employees as the firm’s key asset and a focal source of competitive advantage
and value creation (e.g., Cascio, 1991; Huselid, 1995; Whitener, 2001; Gelade
and Ivery, 2003; Faleye and Trahan, 2011; Edmans, 2012). An extensive
literature has documented that employee-friendly practices and organisational
diversity management policies benefit firms, for instance, by advancing
employee motivation and engagement, labour stability and productivity, and
the firm’s competitiveness in the labour market (e.g., Wright et al., 1995;
Waddock and Graves, 1997; Richard, 2000; Bridges et al., 2003; Jackson et al.,
2003; Kochan et al., 2003; Armstrong et al., 2010; Edmans, 2011, 2012; Chen
et al., 2016; Fauver et al., 2018; Ahmed and Bukth, 2019).
Accordingly, consistent with the stakeholder arguments, intangible invest-
ments in employee welfare and satisfaction may ultimately improve firm
performance by enhancing the firm’s relational and reputational capital with its
employees and other stakeholders. This line of argumentation is also broadly
consistent with the institutional and legitimacy theories used in the CSR
literature in tandem with the stakeholder theory (see, e.g., Fernando and
Lawrence, 2014). These theories suggest that CSR initiatives such as those
related to LGBT-friendly policies can be induced by the firms’ pursuit to build
reputation and achieve legitimacy in the context of their social environment
(e.g., Roumpi et al., 2020). Given that perceived LGBT friendliness is to a large
extent conjoined with inclusive and non-discriminatory employee policies and
3An extensive literature has been devoted to examining the relation between CSR and
firm performance (for reviews, see, e.g., Griffin and Mahon, 1997; Orlitzky, 2001;
Orlitzky et al., 2003; van Beurden and Go¨ssling, 2008; Peloza, 2009; Aguinis and Glavas,
2012; Huang, 2021). While many studies have documented a negative, neutral or mixed
association between CSR activities and financial performance (e.g., McWilliams and
Siegel, 2000; Brammer and Millington, 2008; Surroca and Tribo´, 2008; Surroca et al.,
2010; Kru¨ger, 2015; Buchanan et al., 2018), recent empirical evidence generally supports
the view that the relationship is positive (e.g. Barnett and Salomon, 2012; Servaes and
Tamayo, 2013; Eccles et al., 2014; Dimson et al., 2015; Flammer, 2015; Hasan et al.,
2018; Miller et al., 2020; Huang, 2021).
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 215
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
embracing diversity in the workplace, the concomitant favourable HRM
outcomes are a potential mechanism through which LGBT-friendly corporate
policies can improve firm performance. Nevertheless, analogously to the
stakeholder view, a prerequisite for value creation is that LGBT friendliness
does not alienate the firm’s employees or other stakeholders who have different
social values.
A growing body of research offers evidence that LGBT friendliness advances
a range of desired corporate outcomes. Previous studies have documented that
LGBT-friendly firms are associated with greater employee commitment,
improved job satisfaction, increased employee productivity, and more altruistic
workplace behaviour (e.g., Day and Schoenrade, 1997, 2000; Button, 2001;
Ragins and Cornwell, 2001; Ragins et al., 2007; Badgett et al., 2013; Shan et al.,
2017). Furthermore, LGBT-friendly policies may improve competitiveness in
the labour market by fostering the firm’s ability to attract, recruit and retain
talented employees (e.g., Huffman et al., 2008; Day and Greene, 2008; Metcalf
and Rolfe, 2011; Badgett et al., 2013; Trau, 2015; Wettstein and Baur, 2016).
Among firms’ LGBT workforce, the implementation of sexual minority policies
is associated with improved job satisfaction and psychological well-being, lower
job-related stress and reduced perception of discrimination (e.g., Day and
Schoenrade, 1997; Ragins and Cornwell, 2001; Griffith and Hebl, 2002; Ragins
et al., 2007). LGBT friendliness may also advance customer relations and
improve the firm’s reputation as a socially responsible corporate citizen (Day
and Greene, 2008; Weinzimmer and Esken, 2016; Wettstein and Baur, 2016).
Taken together, the prior literature demonstrates that LGBT-friendly corpo-
rate policies may help firms to accumulate and develop intangibles related to
human capital, stakeholder relations and firm reputation.
Given that LGBT friendliness may enhance the firm’s relational and
reputational capital, it is not surprising that several studies have recently
examined the implications of LGBT-friendly corporate policies on financial
outcomes. Johnston and Malina (2008) and Wang and Schwarz (2010) focus on
the association between firms’ LGBT policies and stock returns. Using an event
study approach, Johnston and Malina (2008) find that the short-run stock
market reaction to news regarding LGBT-friendly policies is positive or
neutral, while Wang and Schwarz (2010) document that firms with more
LGBT-friendly policies are associated with higher long-run stock returns. Li
and Nagar (2013) examine how the adoption of same-sex domestic partner
benefit policies affects stock prices, and report that a portfolio of firms
initiating such LGBT-supportive policies generates excess stock returns of
about 10 percent per year. Chintrakarn et al. (2020) focus on the effects of
LGBT friendliness on credit ratings which directly influence the firm’s
borrowing costs. Their findings demonstrate that more LGBT-friendly firms
have better credit ratings, suggesting that LGBT-friendly policies may pay off
for the firms by reducing the cost of debt and advancing access to external
financing.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
216 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Most directly related to our study, Pichler et al. (2018) and Shan et al. (2017)
investigate the relationship between LGBT-friendly corporate policies and firm
performance. Using data on US firms for the years 1996–2009, Pichler et al.
(2018) document that the impact of LGBT-friendly corporate policies on
profitability and stock market valuation is positive only for firms engaged in
research and development activities. Interestingly, their results indicate that
LGBT-friendly firms without R&D activities are associated with lower
profitability and have no difference in market valuation relative to less
LGBT-friendly firms. Shan et al. (2017) use data on large US firms over the
bullish, pre-crisis period 2002–2006 to examine the influence of LGBT
friendliness on stock returns, stock market valuation and net income per
employee. Their empirical findings demonstrate that firms with more LGBT-
friendly policies have higher risk-adjusted stock returns, higher stock market
valuations, and higher income per employee.
In this paper, we take a new perspective on the effects of LGBT friendliness
on firm performance. Specifically, using data on 657 publicly traded US firms
over the period 2003–2016, we examine how LGBT-friendly corporate policies
influence the firm’s stock market valuation and profitability. Furthermore,
given that the relationship is likely to depend on stakeholders’ socio-political
preferences, we also investigate whether normative social values moderate the
linkage between LGBT-friendly policies and firm performance.
This paper contributes to the literature in three main respects. First, while
our paper builds on the work of Pichler et al. (2018) and Shan et al. (2017), we
aim to provide rigorous new evidence on the relationship between LGBT-
friendly policies and firm performance by utilising a refined empirical approach
which facilitates causal inferences. Second, we extend Pichler et al. (2018) and
Shan et al. (2017) by assessing how stakeholders’ normative social values
influence the relation between LGBT-friendly policies and firm performance.
Although both the stakeholder theory and the HRM theories related to
employee satisfaction and diversity management can be used to hypothesise a
positive association between LGBT friendliness and firm performance, both
mechanisms also suggest that the linkage is likely to depend on stakeholders’
social values and preferences. As noted by Kaplan (2006), Day and Greene
(2008) and Wettstein and Baur (2016), the adoption of LGBT-friendly policies
may lead to stakeholder alienation and backlash if the policies conflict with the
social values of the key stakeholders. Thus, we investigate how social norms
potentially moderate the link between LGBT-friendly policies and firm
performance by exploiting regional differences in social conservatism in our
empirical analysis.4 Finally, given the positive shift in general societal attitude
towards sexual minorities over the past decade, we aim to reinforce the prior
empirical evidence by using a long, more recent sample period which allows us
4Kitzmueller and Shimshack (2012) and Cahan et al. (2020) argue that locally accepted
social norms, views and values may discipline firms into certain social behaviour.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 217
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
to examine how the relationship between LGBT friendliness and firm
performance has evolved over time and under different market conditions.
In our empirical analysis, we follow Shan et al. (2017) and utilise the
Corporate Equality Index (CEI) constructed by the Human Rights Campaign
to measure firm-level LGBT friendliness.5 The CEI is considered to provide a
comprehensive assessment of a firm’s LGBT friendliness in terms of corporate
policies and practices pertaining to LGBT employees as well as public advocacy
related to the rights of sexual minorities. With respect to firm performance, we
use Tobin’s Q as a proxy for stock market valuation and measure firm
profitability with return on assets (ROA). We empirically test the hypothesis
that LGBT-friendly corporate policies are positively associated with firm
performance by using fixed-effects panel regressions in which we control for a
wide variety of firm characteristics including the level of engagement in CSR
activities and overall employee satisfaction. To alleviate endogeneity concerns
and facilitate causal inferences, we utilise two-stage instrumental variable
regressions and propensity score matching in our additional tests. Following
Chintrakarn et al. (2020), our instrument for firm-level LGBT friendliness is the
percentage of the LGBT population in the firm’s headquarter state which
arguably should not have any conceptual relation to the performance of
individual firms. Finally, we exploit regional variation in religiousness and
presidential election results to examine whether and how social norms and
attitudes influence the link between LGBT-friendly policies and firm perfor-
mance.
Our empirical findings demonstrate that LGBT-friendly corporate policies
pay off. Specifically, we document that firms with more LGBT-friendly policies
are more profitable and have higher stock market valuations after controlling
for firm attributes such as size, riskiness, growth and engagement in social
responsibility. The documented positive relationship between LGBT friendli-
ness and firm performance can be considered economically significant; our
estimates suggest that a one standard deviation increase in the firm’s CEI is
associated with an about 7 percent increase in stock market valuation and
approximately 50 basis point increase in ROA.
These results should be compared and contrasted to the partially contradic-
tory evidence provided in the prior studies. While Pichler et al. (2018) do not
find any differences in market valuations between more and less LGBT-friendly
firms unless the firms are engaged in research and development activities, the
findings of Shan et al. (2017) indicate that a one standard deviation increase in
CEI would increase the firm’s market valuation by about 3 percent. Therefore,
our empirical findings extend the earlier evidence by demonstrating an
economically much stronger and statistically highly significant positive
relationship between LGBT-friendly policies and stock market valuation.
5The empirical analysis in Pichler et al. (2018) is based on a dummy variable for firms
that have adopted LGBT-supportive corporate policies.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
218 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Moreover, in contrast to the negative association between LGBT friendliness
and profitability documented in Pichler et al. (2018), our findings provide
considerable evidence to suggest that more LGBT-friendly firms are associated
with significantly higher ROA.6
With respect to the influence of socio-political norms and attitudes towards
sexual minorities, we contribute to the prior literature by documenting that
regional differences in the religious and political leanings moderate the
relationship between LGBT-friendly policies and firm performance. In partic-
ular, our empirical findings demonstrate that the positive effect of progressive
LGBT policies on profitability and market valuation is more pronounced for
firms located in more liberal states while being weaker or non-existent for firms
located in more conservative states. Our results suggest that a standard
deviation increase in the firm’s CEI increases stock market valuation by almost
7 percent in less religious and decisively Democratic states and by about 3
percent in more religious and decisively Republican states. Nevertheless, it is
worthwhile to emphasise that even for firms located in more socially
conservative states, the effect of LGBT friendliness on firm performance is
positive or at worst neutral, suggesting that the adoption of LBGT-friendly
policies does not generally have detrimental repercussions.
2. Data and variables
The sample used in our empirical analysis consists of 657 publicly traded US
firms over the period 2003–2016. Our main analysis requires data on the firms’
(i) LGBT friendliness, (ii) financial statements, (iii) stock prices and (iv)
corporate social responsibility and governance attributes. We measure firm-
level LGBT friendliness with the Corporate Equality Index (CEI) scores
obtained from the Human Rights Campaign. The data on the firms’ income
statement and balance sheet variables, stock prices, and environmental, social
and corporate governance (ESG) scores are obtained from Thomson Reuters.
In our additional tests, we also utilise state-level data on religiousness, US
presidential election results and LGBT demographics. These data are collected
from Gallup, the US National Archives and Records Administration, and the
UCLA Williams Institute, respectively.
We restrict our sample to firms for which the CEI score and the financial data
are available for at least five consecutive years. After excluding penny stocks
and firms with missing data, we obtain an unbalanced panel of 657 individual
6Instead of examining the impact of LGBT-friendly policies on ROA (i.e., the ratio of
net income to total assets), Shan et al. (2017) focus on the ratio of gross income to the
number of employees.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 219
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
firms and approximately 3,000 usable firm-year observations for our main
regressions.7 The final sample includes firms from all major industries that are
headquartered across 42 different US states.
2.1. LGBT friendliness
Following Johnston and Malina (2008), Wang and Schwarz (2010), Everly
and Schwarz (2015) and Shan et al. (2017), we use the Corporate Equality
Index (CEI) constructed by the Human Rights Campaign to measure firm-level
LGBT friendliness. The Human Rights Campaign is the largest LGBT civil
rights advocacy organisation in the US and it has published the CEI for large
US firms annually since 2002. The CEI provides a comprehensive assessment of
a firm’s LGBT friendliness in terms of corporate policies and practices
pertaining to lesbian, gay, bisexual and transgender employees and public
advocacy related to the rights of sexual minorities. The Human Rights
Campaign compiles and constructs the CEI based on self-reported surveys as
well as SEC filings, employee resource groups, press releases and news articles
during the year leading up to the date of publication.8 The surveys underlying
the CEI are sent to the S&P 500 firms, the Fortune 1000 firms, the firms in the
Forbes’ list of 200 largest privately held companies, and other US firms with at
least 500 employees.9 In our empirical analysis, the sample is constrained to
publicly traded firms.
The CEI is based on five main criteria related to firms’ employee policies,
workplace equality, diversity culture and competency, and public statements
and actions related to either advocacy or discrimination of sexual minorities.
The criteria underlying the CEI are summarised in Table 1. Each of the
considered criteria is given a specific amount of points and the CEI is then
constructed for each firm as the sum of the points of the individual evaluation
criteria. Consequently, the CEI may take values between −25 and 100 with
higher values of the index corresponding to more LGBT-friendly corporate
7The number of firm-year observations in our main panel regressions varies from 2,858
to 3,071 depending on the model specification.
8The CEI is released by the Human Rights Campaign during the autumn of each year.
Since 2007, the CEI has been published in a forward-looking manner so that the report
published in the autumn of each year is labelled as the CEI for the upcoming calendar
year. In our empirical analysis, we use the actual year of publication of the CEI
throughout the sample period to maintain conformity (i.e., the 2017 CEI scores
published in the autumn of 2016 are used for calendar year 2016).
9The number of firms covered by the CEI has gradually increased over the sample
period. In 2002, the Human Rights Campaign surveyed the Fortune 500 firms, the firms
in the Forbes’ list of 200 largest privately-held companies, and other US firms with at
least 500 employees.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
220 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
policies and practices. In our empirical analysis, we restrict the sample to
publicly traded firms with non-zero CEI scores.10
2.2. Local socio-political values
We aim to contribute to the prior literature by examining whether local social
values moderate the linkage between LGBT-friendly policies and firm
performance. For this purpose, we use the addresses of firms’ headquarters
and utilise state-level data on religiousness and the US presidential election
results from 2004 to 2016 to divide our sample into subsamples of firms located
in more liberal and more conservative states.
We define a firm to be located in a conservative state if the Republican
candidate won the presidential elections in that state with a margin of at least 5
percent and more than two-thirds of the state population consider themselves
to be ‘highly religious’. Correspondingly, a firm is located in a liberal state if the
Democratic candidate won the presidential elections in that state with a margin
Table 1
The criteria underlying the Corporate Equality Index (CEI)
Criteria 1 Equal Employment Opportunity policies
a) Sexual orientation for all operations 15 points
b) Gender identity for all operations 15 points
c) Contractor/vendor standards include sexual orientation
and gender identity
5 points
Criteria 2 Employment benefits
a) Equivalent spousal and partner benefits 10 points
b) Other ‘soft’ benefits 10 points
c) Transgender-inclusive health insurance coverage 10 points
Criteria 3 Organisational LGBT competency
a) Competency training, resources and accountability measures 10 points
b) Employee group or Diversity council 10 points
Criteria 4 Public commitment
LGBT-specific efforts (recruitment, philanthropy, etc.) 15 points
Criteria 5 Deductions for large-scale anti-LGBT blemish
25-point reduction for recent cases of LGBT discrimination 100 points
10We restrict the sample to firms with non-zero CEI scores in order to alleviate potential
self-selection bias. During the latter half of the sample period, the Human Rights
Campaign has published the CEI also for firms that have not responded to the survey
and have not acquiesced to provide information regarding their employee policies and
the management of sexual orientation diversity. This inconsistency in the requirement
for voluntary disclosure is mainly manifested in the increase in CEI scores of 0 during
the latter half of our sample. In the robustness checks discussed in Section 3.5, we
perform three additional tests to ensure that our results are not influenced by non-
voluntary CEI scores.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 221
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
of at least 5 percent and less than one-third of the state population consider
themselves to be ‘highly religious’. We use a conservative approach to assigning
state-level political stance by requiring a winning margin of at least 5 percent
and by carrying the election results backward over the four years leading up to
the election in order to alleviate concerns that political changes may affect
future corporate policies.11 It may be argued, of course, that the location of the
firm’s headquarters is an imperfect proxy for the social attitude of the firm’s
stakeholders given that most large firms are operating globally. The counter-
argument is that corporate cultures are sticky and firms tend to be entrenched
in social values prevalent in their place of origin. As noted, for example, by
Hilary and Hui (2009), Kitzmueller and Shimshack (2012) and Cahan et al.
(2020), the firm’s environment and locally accepted social norms, views and
values may discipline firms into certain behaviour.
2.3. Firm performance
Following the prior literature (e.g., Huselid, 1995; Waddock and Graves,
1997; Faleye and Trahan, 2011; Servaes and Tamayo, 2013; Chen and Jermias,
2014; Pichler et al., 2018), we measure firm performance with stock market
valuation and profitability. We employ the logarithm of Tobin’s Q as a proxy
for market valuation. Tobin’s Q is calculated as the sum of the firm’s market
value of equity and the book value of liabilities divided by the book value of
total assets. We measure firm profitability with return on assets (ROA) which is
calculated as net income divided by the book value of total assets.
2.4. Control variables
We include a number of control variables in our empirical analysis to account
for the effects of firm-specific factors such as size, financial leverage, growth and
riskiness on firm performance. Specifically, the following set of control
variables is used in the regressions: (i) Size is measured as the logarithm of total
assets, (ii) Leverage is the ratio of total debt to book value of equity, (iii)
Growth is the percentage change in sales from year t – 1 to year t, (iv) Risk is
the firm’s beta coefficient which is estimated against the S&P 500 index using
daily stock return data for year t, (v) ESG is the Thomson Reuters ESG score
which is used as a proxy for the firm’s environmental and social responsibility
and the strength of corporate governance mechanisms, (vi) Board size is the
number of members on the firm’s board of directors, and (vii) Board
independence is measured as the percentage of independent directors on the
board. In the regressions with Tobin’s Q as the dependent variable, we also
include ROA as an additional control variable. These control variables are
11This means that the results of the 2016 Trump vs. Clinton election, for instance, are
utilised for determining state-level political stance for years 2013–2016.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
222 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
selected based on the existing firm performance literature (see, e.g., Capon
et al., 1990; Huselid, 1995; Waddock and Graves, 1997; Aldamen et al., 2012;
Cahan et al., 2015; Frijns et al., 2016; Shan et al., 2017). In addition, we control
for potential biases related to omitted and/or unobservable variables with
industry fixed-effects based on standard industrial classification (SIC) codes
and we account for potential time fixed-effects by including fiscal year dummy
variables in the regressions.
2.5. Instrumental variable
In our additional tests, we address endogeneity concerns with two-stage
instrumental variable regressions. Following Chintrakarn et al. (2020), our
choice of the instrumental variable for firm-level LGBT friendliness is the
percentage of the state population that identifies as lesbian, gay, bisexual or
transgender.
3. Empirical analysis
3.1. Descriptive statistics and correlations
Table 2 reports descriptive statistics for LGBT friendliness (CEI score), our
two alternative dependent variables (Tobin’s Q and ROA), and the control
variables used in the regressions. All variables are presented in their original
forms without logarithms or any other transformations. The mean CEI score
for the firms included in our sample is 68.1 with a standard deviation of 33.7
and the 25th to 75th percentile range from 40 to 100, indicating that the level of
LGBT friendliness varies considerably across firms. Nevertheless, the relatively
high mean CEI score may be indicative of a possible voluntary response bias in
the CEI because firms that have implemented LGBT-friendly corporate policies
or acknowledge the importance of diversity management may be more likely to
respond to the Human Rights Campaign’s survey.12
Table 2 demonstrates that the firms included in our sample exhibit
considerable dispersion in terms of performance. Tobin’s Q, our measure of
market valuation, has a mean value of 2.00 and ranges from 0.82 to 6.98. Firm
profitability, as measured by ROA, varies between −24.87 and 24.71 percent,
with a mean of 6.47 percent. With respect to the control variables, the
descriptive statistics in Table 2 indicate that our sample is very heterogeneous
in terms of firm size, leverage, growth, riskiness and ESG performance.
Table 3 presents the bivariate correlation coefficients between the variables
used in our empirical analysis. As can be seen from the table, CEI score is
significantly positively correlated with Tobin’s Q and ROA. Thus, consistent
12In the robustness checks in Section 3.5, we conduct several additional tests to ensure
that our results are not influenced by biases related to voluntary disclosure.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 223
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
with the hypothesis that LGBT friendliness improves firm performance, the
correlations suggest that firms with more LGBT-friendly corporate policies are
more profitable and have higher stock market valuations. Table 3 further
shows that CEI score is significantly positively correlated with Size, ESG and
Board size, and negatively correlated with Risk, indicating that LGBT-friendly
firms tend to be larger, more socially responsible, less risky and have larger
boards of directors.
As expected, our two firm performance measures Tobin’s Q and ROA are
strongly positively correlated with each other. The firm performance measures
are also statistically significantly correlated with most of our control variables.
Regarding the correlations among the control variables, it can be concluded
from Table 3 that multicollinearity should not be a concern in our regressions
because all the correlation coefficients between the independent variables are
relatively low in magnitude, all being less than 0.4.
Table 2
Descriptive statistics
Variable Mean Median P1 P25 P75 P99 SD
No.
of
obs.
LGBT friendliness
CEI score 68.07 80.00 0.00 40.00 100.00 100.00 33.68 3,123
Firm performance
Tobin’s Q 2.00 1.69 0.82 1.16 2.13 6.98 1.08 3,123
ROA 6.47 6.30 −24.87 1.71 8.69 24.71 7.61 3,123
Control variables
Size 38826.95 13713.33 249.72 3285.61 26415.42 857574.5 98301.93 3,123
Leverage 1.43 0.56 −21.76 0.18 1.00 25.25 25.76 3,123
Growth 4.97 4.35 −42.04 −0.87 14.22 102.06 15.94 3,123
Risk 1.09 1.03 0.09 0.77 1.43 3.02 0.51 3,123
ESG 46.58 43.07 17.13 33.81 52.93 84.94 15.97 3,123
Board size 11.07 11.00 6.00 9.00 12.00 18.00 2.06 3,123
Board
independence
81.79 84.62 33.33 75.00 90.00 93.75 11.51 3,123
The table reports summary statistics for the sample of 657 publicly traded US firms over the
period 2003–2016. LGBT friendliness is measured with the Corporate Equality Index (CEI
score) constructed by the Human Rights Campaign. The dependent variables are defined as
follows: Tobin’s Q is the sum of the firm’s market value of equity and the book value of
liabilities divided by the book value of total assets and ROA is calculated as net income
divided by the book value of total assets. The control variables are defined as follows: Size is
measured with the firm’s total assets, Leverage is the ratio of total debt to book value of
equity, Growth is the percentage change in sales from year t – 1 to year t, Risk is the firm’s
beta coefficient, ESG is the Thomson Reuters Environmental, Social and Governance
responsibility score, Board size is the number of members on the firm’s board of directors,
and Board independence is the percentage of independent directors on the board.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
224 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
3.2. Main results
We test our main research hypothesis that LGBT-friendly corporate policies
are positively associated with firm performance by estimating alternative
versions of the following panel regression specification:
Performancei,t ¼ αþβCEI scorei,tþ γ Firm–specific controlsð Þi,t
þω Industry fixed–effectsð Þi,tþφ Year fixed–effectsð Þi,tþ ɛi,t
(1)
where the dependent variable Performancei,t is one of the two alternative firm
performance measures (Tobin’s Q or ROA) for firm i at time t and CEI scorei,t is
the Corporate Equality Index for firm i at time t which is our proxy for firm-
level LGBT friendliness. The firm-specific control variables in Equation (1) are
Size, Leverage, Growth, Risk, ESG, Board size and Board independence. In the
regressions with Tobin’s Q as the dependent variable, we also include ROA as
Table 3
Correlations
(1) (2) (3) (4) (5) (6) (7) (8) (9)
(1) CEI score
(2) Tobin’s Q 0.158*
(3) ROA 0.100* 0.613*
(4) Size 0.240* −0.206* −0.099*
(5) Leverage −0.032 −0.069* −0.165* 0.006
(6) Growth −0.014 0.214* 0.210* 0.012 −0.036
(7) Risk −0.164* −0.260* −0.338* −0.097* 0.064* −0.047
(8) ESG 0.104* 0.055* 0.099* −0.015 −0.027 −0.064* −0.019
(9) Board size 0.190* −0.058* 0.029 0.388* 0.004 −0.024 −0.167* 0.031
(10) Board
independence
0.035 −0.023 0.022 0.116* −0.006 −0.077* −0.034 0.251* 0.062*
The table reports pairwise correlations between the variables used in the main regressions.
LGBT friendliness is measured with the Corporate Equality Index (CEI score) constructed by
the Human Rights Campaign. The dependent variables are defined as follows: Tobin’s Q is
the sum of the firm’s market value of equity and the book value of liabilities divided by the
book value of total assets and ROA is calculated as net income divided by the book value of
total assets. The control variables are defined as follows: Size is measured as the logarithm of
the firm’s total assets, Leverage is the ratio of total debt to book value of equity, Growth is the
percentage change in sales from year t – 1 to year t, Risk is the firm’s beta coefficient, ESG is
the Thomson Reuters Environmental, Social and Governance responsibility score, Board size
is the number of members on the firm’s board of directors, and Board independence is the
percentage of independent directors on the board. All variables are winsorised at the 1st and
99th percentiles. * denotes statistical significance at the 0.01 level.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 225
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
an additional control variable. This set of control variables should account for
the potentially confounding effects of firm characteristics on profitability and
market valuation. Equation (1) also includes industry fixed-effects to control
for systemic variation in firm performance across different industries as well as
potential biases related to omitted variables and unobserved heterogeneity.
Moreover, we account for systematic variation in firm performance over time
by including year fixed-effects in the regressions. All variables in Equation (1)
are winsorised at the 1st and 99th percentiles to moderate the effects of outliers.
Throughout the estimations, we use robust standard errors that are adjusted
for heteroskedasticity and clustered by firm.
The estimation results of four alternative versions of Equation (1) are
reported in Table 4. The dependent variable is Tobin’s Q in Models 1 and 2 and
ROA in Models 3 and 4. Furthermore, Models 1 and 3 are baseline regressions
without industry and year fixed-effects and with a constrained set of control
variables (Models 1 and 3), whereas Models 2 and 4 include the full set of firm-
specific controls as well as industry and year fixed-effects. As shown in Table 4,
the F-statistics are statistically significant at the 1 percent level in all four model
specifications, and the adjusted R2s indicate that our panel regressions explain
about 50 percent of the variation in Tobin’s Q and about 20 percent of the
variation in ROA.
Overall, the regression results in Table 4 indicate that LGBT friendliness is
positively associated with firm performance. The coefficient estimates for CEI
score are positive and statistically significant at the 1 percent level in every
model specification, suggesting that firms with more LGBT-friendly corporate
policies are more profitable and have higher stock market valuations. In
addition to being statistically highly significant, the positive relationship
between LGBT friendliness and firm performance can also be considered
economically significant. The magnitudes of the estimated coefficients suggest
that a one standard deviation increase in CEI score would increase the firm’s
market valuation by approximately 7 percent and return on assets by about 47
to 51 basis points. Overall, the estimates reported in Table 4 provide strong
support for our hypothesis that LGBT-friendly corporate policies improve firm
performance.
With respect to stock market valuation, our regression results in Table 4
should be compared and contrasted to the findings of Shan et al. (2017) and
Pichler et al. (2018). Using the Corporate Equality Index over the period
2002–2006, Shan et al. (2017) document that a ten-point increase in CEI score
would be associated with an approximately 1 percent increase in stock market
valuation. On the other hand, Pichler et al. (2018), who base their analysis on a
dummy variable for firms with LGBT-supportive corporate policies and use
MSCI ESG data over the period 1996–2009, do not find any significant
differences in market valuations between more and less LGBT-friendly firms
unless the firms are engaged in research and development activities. Therefore,
our empirical findings extend the earlier results of Shan et al. (2017) and Pichler
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
226 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 4
Regression results
Tobin’s Q ROA
Model 1 Model 2 Model 3 Model 4
Constant 0.921*** 0.976*** 15.536*** 14.026***
(9.78) (8.03) (11.14) (7.45)
CEI score 0.002*** 0.002*** 0.014*** 0.015***
(5.14) (5.15) (2.66) (2.77)
Size −0.067*** −0.071*** −0.825*** −0.792***
(−7.52) (−6.92) (−5.98) (−4.77)
Leverage 0.003 0.003 −0.219*** −0.211***
(1.53) (1.33) (−4.59) (−4.36)
ROA 0.039*** 0.038***
(16.86) (16.25)
Growth 0.003*** 0.003*** 0.074*** 0.073***
(3.18) (3.56) (7.08) (6.72)
Risk −0.099*** −0.091*** −3.698*** −3.295***
(−5.91) (−5.03) (−11.54) (−8.98)
ESG 0.000 0.000 0.038*** 0.033***
(−0.07) (−0.65) (3.92) (3.30)
Board size 0.000 0.082
(0.13) (0.85)
Board independence 0.000 0.001
(−0.44) (0.08)
Industry fixed-effects No Yes No Yes
Period fixed-effects No Yes No Yes
No. of observations 2,868 2,858 3,071 3,060
Adjusted R2 0.49 0.53 0.20 0.22
F-stat. 388.19*** 122.66*** 132.82*** 36.39***
The table reports the estimates of four alternative versions of Equation (1). LGBT
friendliness is measured with the Corporate Equality Index (CEI score) constructed by the
Human Rights Campaign. The dependent variables are defined as follows: Tobin’s Q is the
logarithm of the sum of the firm’s market value of equity and the book value of liabilities
divided by the book value of total assets and ROA is calculated as net income divided by the
book value of total assets. The control variables are defined as follows: Size is measured as the
logarithm of the firm’s total assets, Leverage is the ratio of total debt to book value of equity,
Growth is the percentage change in sales from year t – 1 to year t, Risk is the firm’s beta
coefficient, ESG is the Thomson Reuters Environmental, Social and Governance responsi-
bility score, Board size is the logarithm of the number of members on the firm’s board of
directors, and Board independence is the percentage of independent directors on the board. All
variables are winsorised at the 1st and 99th percentiles. The t-statistics (in parentheses) are
based on robust standard errors which are adjusted for heteroskedasticity and are clustered
by firm. ***, ** and * denote statistical significance at the 0.01, 0.05 and 0.10 levels,
respectively.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 227
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
et al. (2018) by demonstrating that LGBT-friendly corporate policies are
strongly positively associated with stock market valuation over the period
2003–2016.
While Shan et al. (2017) do not examine the relationship between LGBT
friendliness and firm profitability, Pichler et al. (2018) document that more
LGBT-friendly firms without engagement in R&D activities are associated with
significantly lower ROA. However, their findings also indicate that LGBT-
friendly firms with R&D activities are more profitable and have an approx-
imately one percentage point higher ROA. Intriguingly, in stark contrast to the
negative association between LGBT friendliness and profitability documented
in Pichler et al. (2018), our results reported in Table 4 provide strong evidence
to suggest that firms with more LGBT-friendly corporate policies have higher
profitability. These contrasting findings with respect to firm profitability may
be at least partially reconciled by considering the differences in the LGBT
friendliness proxies (CEI score vs. MSCI ESG dummy) and the sample periods
(2003–2016 vs. 1996–2009) used in the respective empirical analyses.
As can be noted from Table 4, the coefficient estimates for most of our
control variables are statistically highly significant, demonstrating the impor-
tance of these variables as determinants of firm market valuation and
profitability. Specifically, the regression results suggest that firm performance
is negatively associated with Size, Leverage and Risk, while being significantly
positively related to Growth and ESG. Furthermore, as expected, the
regressions with Tobin’s Q as the dependent variable indicate that stock
market valuation is strongly positively associated with firm profitability
(p < 0.01).
Although both the stakeholder theory and the HRM theories related to
employee satisfaction and diversity management can be used to hypothesise a
positive association between LGBT friendliness and firm performance, both
mechanisms also suggest that the linkage is likely to depend on stakeholders’
values and preferences. Therefore, we next investigate whether and how
regional differences in conservatism and the general social attitude towards
sexual minorities potentially moderate the link between LGBT-friendly policies
and firm performance. For this purpose, we use the addresses of the firms’
headquarters and state-level data on religiousness and the US presidential
election results to split our sample into subsamples of firms located in more
liberal and more conservative states. We then estimate alternative versions of
Equation (1) using the two subsamples.
Table 5 reports the estimation results of Equation (1) based on the
subsamples of firms located in more liberal and more conservative states.
Similar to Table 4, Tobin’s Q is used as the dependent variable in Models 1 and
2 and ROA in Models 3 and 4. All four regression specifications in Table 5
include the full set of firm-specific control variables and account for industry
and year fixed-effects. As can be seen from the table, the F-statistics are
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
228 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 5
The influence of social norms on the relation between LGBT friendliness and firm performance
Tobin’s Q ROA
Liberal Conservative Liberal Conservative
Constant 0.817*** 0.605* −11.085*** 9.685***
(3.81) (1.82) (−2.72) (2.68)
CEI score 0.002** 0.001* 0.030* 0.018
(2.14) (1.65) (1.93) (1.64)
Size −0.041** −0.031 −1.063*** −0.764**
(−2.01) (−1.26) (−2.89) (−2.44)
Leverage 0.007 0.006 −0.176 −0.222***
(1.12) (1.44) (−1.03) (−3.05)
ROA 0.044*** 0.036***
(9.04) (8.17)
Growth 0.006** −0.000 0.078*** 0.074***
(2.41) (−0.12) (3.36) (3.65)
Risk −0.077* −0.092*** −3.128*** −2.867***
(−1.85) (−2.62) (−3.84) (−5.09)
ESG −0.000 −0.000 0.053*** 0.046**
(−0.32) (−0.09) (2.77) (2.06)
Board size −0.009 0.009 0.103 0.383***
(−0.64) (0.77) (0.44) (2.96)
Board independence −0.000 −0.000 0.034 0.003
(−0.14) (−0.22) (1.06) (0.14)
Industry fixed-effects Yes Yes Yes Yes
Period fixed-effects Yes Yes Yes Yes
No. of observations 552 609 597 624
Adjusted R2 0.62 0.45 0.26 0.26
The table reports the estimates of four alternative versions of Equation (1) based on subsamples
of firms located in more liberal and more conservative states. A firm is located in a conservative
state if the Republican candidate won the presidential elections with a margin of at least 5
percent and more than two-thirds of the state population consider themselves to be highly
religious. Correspondingly, a firm is located in a liberal state if the Democratic candidate won
the presidential elections with a margin of at least 5 percent and less than one-third of the state
population consider themselves to be highly religious. LGBT friendliness is measured with the
Corporate Equality Index (CEI score) constructed by the Human Rights Campaign. The
dependent variables are defined as follows: Tobin’s Q is the logarithm of the sum of the firm’s
market value of equity and the book value of liabilities divided by the book value of total assets
and ROA is calculated as net income divided by the book value of total assets. The control
variables are defined as follows: Size is measured as the logarithm of the firm’s total assets,
Leverage is the ratio of total debt to book value of equity, Growth is the percentage change in
sales from year t – 1 to year t, Risk is the firm’s beta coefficient, ESG is the Thomson Reuters
Environmental, Social and Governance responsibility score, Board size is the logarithm of the
number ofmembers on the firm’s boardof directors, andBoard independence is the percentage of
independent directors on the board. All variables are winsorised at the 1st and 99th percentiles.
The t-statistics (in parentheses) are based on robust standard errors which are adjusted for
heteroskedasticity and are clustered by firm. ***, ** and * denote statistical significance at the
0.01, 0.05 and 0.10 levels, respectively.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 229
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
significant at the 1 percent level in every model and the adjusted R2s indicate a
relatively good fit of the estimated regressions.
The estimates in Table 5 demonstrate that the positive association between
LGBT-friendly corporate policies and firm performance is stronger for firms
located in more liberal (i.e., less religious and decisively Democratic) states
while being weaker or non-existent for firms located in more conservative (i.e.,
more religious and decisively Republican) states. Specifically, in both subsam-
ples, the estimated coefficients for CEI score are positive and statistically
significant in the regressions with Tobin’s Q as the dependent variable,
indicating that LGBT-friendly firms have higher market valuations. However,
for firms located in more liberal states, the coefficient estimate of CEI score is
larger in magnitude as well as being statistically more significant. The estimates
of Models 1 and 2 in Table 5 suggest that a one standard deviation increase in
CEI score for firms located in more liberal states increases their stock market
valuation by approximately 7 percent, whereas a corresponding increase for
firms in more conservative states is about 3 percent.
In the regressions with ROA as the dependent variable, the estimated
coefficient of CEI score is positive and statistically significant for firms located
in more liberal states and insignificant at conventional levels (p = 0.102) for
firms headquartered in more conservative states. Thus, our findings suggest
that the positive association between LGBT-friendly corporate policies and
profitability pertains more to firms that are located in less religious and
decisively Democratic states. For these firms, a ten-point increase in CEI score
would be associated with an about 30 basis points increase in ROA. Consistent
with the regressions reported in Table 4, the coefficient estimates for the control
variables demonstrate that firm performance is significantly negatively asso-
ciated with Size, Leverage and Risk, while being positively related to Growth,
ESG and Board size.
In general, the regression results presented in Tables 4 and 5 provide strong
evidence that firms with more LGBT-friendly corporate policies are associated
with higher profitability and higher stock market valuation. Our empirical
findings further demonstrate that the positive effect of progressive LGBT
policies is more pronounced for firms that are headquartered in less religious
and decisively Democratic states and is weaker or non-existent for firms located
in more religious and decisively Republican states. This suggests that regional
differences in political and religious leanings moderate the relationship between
LGBT-friendly policies and firm performance.
3.3. Instrumental variable regressions and propensity score matching
Given that our research hypothesis implies that LGBT-friendly corporate
policies improve firm performance, it is important to acknowledge that our
preceding analysis may suffer from endogeneity and reverse causality. In our
panel regressions, we have controlled for a number of firm-specific
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
230 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
characteristics that are known to affect profitability and market valuation.
Moreover, we have attempted to control for potential biases related to omitted
correlated variables that may influence firm performance by including industry
and year fixed-effects in the regressions. Nevertheless, it is possible that some
unobservable or omitted firm attributes simultaneously affect firm performance
and the implementation of LGBT-friendly policies. It is also plausible that
firms that have better financial performance can allocate more resources to
employee relations and societal activities, potentially leading to reverse
causality from firm performance to LGBT friendliness. As the next step of
our analysis, we utilise two-stage instrumental variable regressions and
propensity score matching in order to alleviate endogeneity concerns and to
establish a causal link between LGBT-friendly corporate policies and firm
performance.
First, we address endogeneity concerns by estimating two-stage instrumental
variable regressions. Following Chintrakarn et al. (2020), our choice of the
instrumental variable for firm-level LGBT friendliness is the percentage of the
state population that identifies as lesbian, gay, bisexual or transgender. We
posit that the percentage of the LGBT population in a given state should be
positively related to the implementation of LGBT-friendly corporate policies,
while it arguably should not have any conceptual relation to the performance
of individual firms. Accordingly, in the first-stage regression, we model CEI
score as a function of LGBT population and the set of control variables used in
Equation (1). In the second-stage regressions, Tobin’s Q and ROA are regressed
on the instrumented CEI score and our firm-specific control variables.
The estimates of the two-stage instrumental variable regressions are
presented in Table 6. As can be seen from the table, the estimated coefficient
for the instrumental variable LGBT population is positive and statistically
highly significant (t-stat = 4.08) in the first-stage regression with CEI score as
the dependent variable. This demonstrates that our instrument is strongly
positively associated with the implementation of LGBT-friendly corporate
policies. Furthermore, the high partial F-statistics of the first-stage regressions
as well as the LM test for underidentification and the Wald test for weak
identification all indicate that LGBT population is a valid instrument for CEI
score, thereby suggesting that our instrumental variable estimates should not be
plagued by a weak-instrument problem. The first-stage regressions in Table 6
also indicate that LGBT friendliness is positively associated with Size, ESG and
Board size and negatively associated with Risk.
The estimates of the second-stage regressions with the instrumented CEI
score are very similar to the results reported in Table 4. Most importantly, the
coefficient estimates for the instrumented CEI score are positive and statisti-
cally highly significant both in the Tobin’s Q and ROA regressions, suggesting
that more LGBT-friendly firms have higher profitability and stock market
valuation even after controlling for potential endogeneity. Therefore, the
instrumental variable regressions in Table 6 provide support for the hypothesis
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 231
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
that LGBT-friendly corporate policies improve firm performance. With respect
to the control variables, the estimates of the second-stage regressions are
consistent with our main analysis with the only exception being the insignificant
coefficients for ESG in both second-stage models.
We aim to further mitigate endogeneity concerns by utilising propensity score
matching. Specifically, we use all the control variables included in Equation (1)
to estimate propensity scores for the sample firms and then use these scores to
identify a matched sample of less LGBT-friendly firms that are statistically
indistinguishable from the most LGBT-friendly firms in terms of size, riskiness,
ESG engagement, and other firm-specific controls. If the only observable
difference between the matched firms is their CEI score, there should
presumably not be any differences in Tobin’s Q and ROA unless firm
performance is affected by LGBT friendliness. We utilise one-to-one nearest
neighbour matching without replacement and require that the maximum
Table 6
Instrumental variable regressions
Variable
First-stage regression
Second-stage regressions
CEI score Tobin’s Q ROA
Constant −36.460** 0.977*** 13.730***
(−2.36) (7.17) (6.61)
Instrumental variables
LGBT population 10.233***
(4.16)
Independent variables
Instrumented CEI score 0.006*** 0.074**
(2.86) (2.33)
Size 4.680*** −0.090*** −1.073***
(4.76) (−6.54) (−4.62)
Leverage −0.150 0.003 −0.198***
(−0.58) (1.33) (−4.15)
ROA 0.037***
(14.28)
Growth −0.046 0.003*** 0.075***
(−1.09) (3.82) (6.71)
Risk −5.853*** −0.070*** −2.961***
(−3.52) (−3.01) (−7.12)
ESG 0.212*** −0.001 0.018
(3.90) (−1.56) (1.46)
Board size 1.257** −0.005 −0.001
(2.35) (−0.66) (−0.01)
Board independence 0.079 −0.000 −0.000
(0.79) (−0.51) (−0.04)
Industry fixed-effects Yes Yes Yes
(continued)
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
232 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
difference between the propensity score of each treatment firm and that of its
matched control firm does not exceed 0.1 standard deviations. After identifying
matching firms for the most LGBT-friendly firms, we re-estimate alternative
versions of Equation (1) using the propensity score matched sample of firms.
Table 7 reports matching diagnostics and the regression results based on the
propensity score matched sample. In order to ascertain that the matched firms
are sufficiently similar to the treatment firms, we first re-estimate the probit
model underlying the propensity score matching using the matched-firm
sample. The pseudo R2 of the post-matching probit model is about 54 percent
lower than the pre-matching pseudo R2 and the post-matching coefficient
estimates for all control variables except for Size and ESG become statistically
insignificant. The mean and the mean percentage differences between the
propensity scores of the treatment and matched firms are 0.01 and 3.9 percent,
respectively. Moreover, the sample means of Size, Leverage, Risk and ESG are
almost equal for the treatment and matched firms. Therefore, we conclude that
the propensity score matching effectively eliminates the observable differences
Table 6 (continued)
Variable
First-stage regression
Second-stage regressions
CEI score Tobin’s Q ROA
Period fixed-effects Yes Yes Yes
No. of observations 3,060 2,858 3,060
Adjusted R2 0.23 0.45 0.14
F-stat. 36.05*** 118.34*** 36.42***
Partial F-stat. 16.73***
LM stat. 24.58*** 24.18***
Wald F-stat. 134.16*** 131.59***
The table reports the estimates of two-stage instrumental variable regressions. LGBT
friendliness is measured with the Corporate Equality Index (CEI score) constructed by the
Human Rights Campaign. The instrumental variable for CEI score is the percentage of the
state population that identifies as lesbian, gay, bisexual or transgender. The dependent
variables are defined as follows: Tobin’s Q is the logarithm of the sum of the firm’s market
value of equity and the book value of liabilities divided by the book value of total assets and
ROA is calculated as net income divided by the book value of total assets. The control
variables are defined as follows: Size is measured as the logarithm of the firm’s total assets,
Leverage is the ratio of total debt to book value of equity, Growth is the percentage change in
sales from year t – 1 to year t, Risk is the firm’s beta coefficient, ESG is the Thomson Reuters
Environmental, Social and Governance responsibility score, Board size is the logarithm of the
number of members on the firm’s board of directors, and Board independence is the
percentage of independent directors on the board. All variables are winsorised at the 1st and
99th percentiles. The t-statistics (in parentheses) are based on robust standard errors which
are adjusted for heteroskedasticity and are clustered by firm. ***, ** and * denote statistical
significance at the 0.01, 0.05 and 0.10 levels, respectively.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 233
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
between the most LGBT-friendly firms and their matched less LGBT-friendly
counterparts.
Overall, the regression results based on the propensity score matched sample
are very similar to our main regressions, and thereby provide further evidence
to suggest that LGBT-friendly corporate policies improve firm performance.
Similar to Tables 5 and 6, Tobin’s Q is used as the dependent variable in
Models 1 and 2 and ROA in Models 3 and 4, and regressions for both
dependent variables are estimated without and with industry and year fixed-
effects. As can be noted from Table 7, the coefficients for CEI score are positive
and statistically highly significant in all four models, indicating that firms with
more LGBT-friendly policies are more profitable and have higher stock market
valuation even when a propensity score matched sample is used in the
regressions. Broadly consistent with our main analysis, the fixed-effects
Table 7
Propensity score matching
Variable
Tobin’s Q ROA
Model 1 Model 2 Model 3 Model 4
Constant 0.884*** 0.938*** 7.288** 7.038**
(4.99) (5.17) (2.57) (2.27)
CEI score 0.001*** 0.002*** 0.019** 0.020**
(2.71) (3.58) (2.42) (2.45)
Size −0.066*** −0.066*** −0.659*** −0.701***
(−5.19) (−5.23) (−3.24) (−3.30)
Leverage 0.007** 0.005 −0.225** −0.213**
(2.07) (1.45) (−2.52) (−2.51)
ROA 0.040*** 0.039***
(12.03) (11.82)
Growth 0.003*** 0.004*** 0.070*** 0.064***
(3.21) (3.98) (4.27) (3.78)
Risk −0.103*** −0.103*** −3.282*** −3.056***
(−4.22) (−4.12) (−6.71) (−6.06)
ESG 0.000 −0.001 0.059*** 0.054***
(0.35) (−0.79) (4.06) (3.50)
Board size 0.001 0.007 0.150 0.148
(0.15) (1.08) (1.23) (1.20)
Board independence 0.000 −0.001 0.038* 0.031
(0.25) (−0.47) (1.95) (1.54)
Industry fixed-effects No Yes No Yes
Period fixed-effects No Yes No Yes
No. of observations 1,246 1,246 1,246 1,246
Adjusted R2 0.47 0.54 0.17 0.20
F-stat. 40.24*** 31.89*** 15.960*** 7.820***
(continued)
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
234 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
estimates in Table 7 suggest that a one standard deviation increase in CEI score
would increase the firm’s market valuation by approximately 7 percent and
return on assets by about 67 basis points.
3.4. Additional tests
We examine the robustness of our results by conducting a number of
additional tests.13 First, even though we have controlled for industry fixed-
effects in our main regressions, we acknowledge that our findings may be
influenced by cross-industry differences in social progressiveness and attitude
Table 7 (continued)
Variable
Tobin’s Q ROA
Model 1 Model 2 Model 3 Model 4
PSM diagnostics
Pre-matching pseudo R2 0.13
Pre-matching LR chi-square 456.38***
Post-matching pseudo R2 0.06
Post-matching LR chi-square 75.11***
Mean difference 0.010
Max difference 0.020
Mean percentage difference 0.039
Max percentage difference 1.646
The table reports the estimates of four alternative versions of Equation (1) based on a
propensity score matched sample of firms. LGBT friendliness is measured with the Corporate
Equality Index (CEI score) constructed by the Human Rights Campaign. We utilise
propensity score matching to build a matched-firm sample in which the most LGBT-friendly
firms with a CEI score of 100 are matched with less LGBT-friendly firms which are as similar
as possible in terms of the control variables. The dependent variables are defined as follows:
Tobin’s Q is the logarithm of the sum of the firm’s market value of equity and the book value
of liabilities divided by the book value of total assets and ROA is calculated as net income
divided by the book value of total assets. The control variables are defined as follows: Size is
measured as the logarithm of the firm’s total assets, Leverage is the ratio of total debt to book
value of equity, Growth is the percentage change in sales from year t – 1 to year t, Risk is the
firm’s beta coefficient, ESG is the Thomson Reuters Environmental, Social and Governance
responsibility score, Board size is the logarithm of the number of members on the firm’s board
of directors, and Board independence is the percentage of independent directors on the board.
All variables are winsorised at the 1st and 99th percentiles. The t-statistics (in parentheses) are
based on robust standard errors which are adjusted for heteroskedasticity and are clustered
by firm. ***, ** and * denote statistical significance at the 0.01, 0.05 and 0.10 levels,
respectively.
13For brevity, we do not tabulate our additional tests. The results of these robustness
checks are available from the corresponding author.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 235
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
towards sexual minorities. These differences are also reflected in the mean CEI
scores in our sample which vary considerably across industries, being highest in
business equipment and lowest in the energy, oil and gas industry. To address
potential industry effects, we estimate industry-adjusted CEI scores for each
firm as the residual from a regression of CEI score on industry dummies. We
then re-estimate alternative versions of Equation (1) using the industry-
adjusted CEI scores as the test variable. The regression results (not tabulated)
are very similar to the results reported in Table 4. Most importantly, the
coefficients for the industry-adjusted CEI score are positive and statistically
significant at the 1 percent level in all four regression specifications. This
suggests that our empirical findings should not be driven by cross-industry
differences in LGBT friendliness.
Second, another potential concern with the CEI score is that it is based
largely on annual self-reported surveys. Since 2011, the Human Rights
Campaign has published the CEI also for firms that have not responded to
the survey, thereby inflicting an inconsistency in the constituent firms. We
conduct three additional tests to investigate whether our findings are influenced
by the non-voluntary CEI scores. We first re-estimate the regressions using two
subsamples; the first subsample excludes all firms with non-voluntary CEI
scores and the second subsample comprises only the firms with non-voluntary
CEI scores. The estimation results for both subsamples (not tabulated) are
remarkably similar and the coefficients for CEI score are positive and
statistically significant at the 1 percent level in all regressions. We also estimate
regressions in which a dummy variable for non-voluntary CEI scores is used as
an additional control variable. Once again, the coefficient estimates (not
tabulated) for CEI score are positive and statistically highly significant
throughout the alternative regressions. Our third approach is to use propensity
score matching to construct a matched-firm sample in which firms with non-
voluntary CEI scores are matched with essentially identical firms that have
responded to the surveys. When the regressions are re-estimated using the
propensity score matched sample, the coefficients for CEI score remain positive
and statistically significant (not tabulated). Thus, we conclude that our results
are robust to alternative approaches to account for potential non-voluntary
disclosure bias in the CEI score.
Third, given that LGBT-friendly corporate policies can be considered as one
dimension of corporate social responsibility policies and diversity management,
we aim to further ascertain that LGBT friendliness has an incremental effect on
firm performance over and above the level of engagement in social responsi-
bility. For this purpose, we replace the ESG score in our regressions with the
Thomson Reuters social responsibility score (S score) which takes values
between 0 and 100 with higher values indicating higher levels of social
responsibility. The regressions results (not tabulated) are consistent with our
main analysis; the coefficient estimates for CEI score are positive and
statistically significant throughout the alternative model specifications. We
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
236 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
also re-estimate the regressions using two subsamples from which either the
most socially responsible firms (S score in the highest decile) or the least socially
responsible firms (S score in the lowest decile) are excluded. Again, the
coefficient estimates (not tabulated) for CEI score are positive and statistically
significant in both subsamples, suggesting that LGBT friendliness is positively
associated with firm performance regardless of the degree of social responsi-
bility.
Fourth, because corporate governance may have confounding effects on the
relation between LGBT-friendly policies and firm performance, we estimate
regressions in which we employ alternative control variables for corporate
governance quality as a further sensitivity test. While in our main regressions
we have included ESG, Board size and Board independence to control for
governance mechanisms, we now estimate regressions in which the ESG score
is replaced with the Thomson Reuters corporate governance score (G score)
and board reputation is included as an additional control variable related to
board characteristics. The G score takes values between 0 and 100 with higher
values indicating stronger corporate governance mechanisms. Following
Unsal and Brodmann (2020), we use the change in the number of outside
board seats held by the firms’ directors as a proxy for board reputation.
Consistent with our main regressions reported in Table 4, the coefficient
estimates for CEI score are positive and remain highly significant in all model
specifications after the inclusion of the additional governance quality control
variables.
Finally, we acknowledge that firms with employee-supportive working
environment and good employee relations are also likely to be more LGBT-
friendly. To ensure that LGBT friendliness has an incremental effect on firm
performance over overall employee friendliness, we next utilise Fortune’s list
of the 100 Best Companies to Work for in America to identify firms with the
highest employee satisfaction.14 We then re-estimate our main regressions
using a sample from which the most employee-friendly firms included in
Fortune’s best employer list have been excluded. As an alternative approach,
we also estimate regressions in which a dummy variable for the most
employee-friendly firms is used as an additional control variable. Irrespective
of the approach, the estimated coefficients for CEI score (not tabulated) are
positive and statistically significant both in the Tobin’s Q and ROA
regressions. Overall, these additional tests suggest that the documented
positive relationship between LGBT friendliness and firm performance is
independent of employee-supportive corporate policies and overall employee
satisfaction.
14Fortune’s list of the 100 Best Companies to Work for in America has been utilised
previously to measure employee satisfaction, for example in Filbeck and Preece (2003),
Edmans (2011, 2012), Faleye and Trahan (2011) and Chen et al. (2016).
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 237
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4. Conclusions
In this paper, we examine the association between LGBT-friendly corporate
policies and firm performance. Given that the relationship is likely to depend
on stakeholders’ socio-political preferences, we further investigate whether
normative social values moderate the linkage between LGBT-friendly policies
and firm performance. We empirically test the hypothesis that LGBT-friendly
corporate policies improve firm performance using data on 657 publicly traded
US firms over the period 2003–2016. In our analysis, we utilise the Corporate
Equality Index (CEI) constructed by the Human Rights Campaign to measure
firm-level LGBT friendliness, and we use stock market valuation and
profitability to measure firm performance.
Consistent with our research hypothesis, we document that firms with more
LGBT-friendly corporate policies have higher profitability and higher stock
market valuations after controlling for firm attributes such as size, riskiness,
growth and overall engagement in social responsibility. The documented
positive relationship between LGBT friendliness and firm performance can be
considered economically significant; our estimates suggest that a one standard
deviation increase in the firm’s CEI is associated with an almost 7 percent
increase in stock market valuation and about 50 basis point increase in
profitability.
Although both the stakeholder theory and the HRM arguments can be used
to hypothesise a positive association between LGBT friendliness and firm
performance, both mechanisms also suggest that the linkage is likely to depend
on stakeholders’ preferences and socio-political values. Thus, we investigate
how social norms and attitudes potentially moderate the linkage between
LGBT-friendly policies and firm performance by exploiting regional differences
in social conservatism in our analysis. Our findings indicate that the positive
effect of progressive LGBT policies on profitability and market valuation is
more pronounced for firms located in more liberal states while being weaker or
non-existent for firms located in more conservative states.
Taken as a whole, our empirical findings provide strong evidence to suggest that
LGBT-friendly corporate policies enhance firm performance. These findings can
be considered to support the view that socially progressive corporate policies and
diversity management pay off and create value for the firm.
References
Agle, B. R., T. Donaldson, R. E. Freeman, M. C. Jensen, R. K. Mitchell, and D. J.
Wood, 2008, Dialogue: toward superior stakeholder theory, Business Ethics Quarterly
18, 153–190.
Aguinis, H., and A. Glavas, 2012, What we know and don’t know about corporate
social responsibility: a review and research agenda, Journal of Management 38,
932–968.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
238 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Ahmed, H. L., and T. Bukth, 2019, Effect of employee diversity on firm performance in
Bangladesh, Frontiers in Management Research 3, 26–39.
Aldamen, H., K. Duncan, S. Kelly, R. McNamara, and S. Nagel, 2012, Audit committee
characteristics and firm performance during the global financial crisis, Accounting and
Finance 52, 971–1000.
Armstrong, C., P. C. Flood, J. P. Guthrie, W. Liu, S. MacCurtain, and T. Mkamwa,
2010, The impact of diversity and equality management on firm performance: beyond
high performance work systems, Human Resource Management 49, 977–998.
Artiach, T., D. Lee, D. Nelson, and J. Walker, 2010, The determinants of corporate
sustainability performance, Accounting and Finance 50, 31–51.
Aspan, M., 2020, Fortune 500 CEOs praise landmark LGBTQ antidiscrimination
ruling. Fortune (16 June). Available at: https://fortune.com/2020/06/16/fortune-500-
ceos-supreme-court-lgbtq-ruling/
Badgett, M. V. L., L. Durso, A. Kastanis, and C. Mallory, 2013, The business impact of
LGBT-supportive workplace policies, Working paper (The Williams Institute,
UCLA).
Barnett, M. L., 2007, Stakeholder influence capacity and the variability of financial
returns to corporate social responsibility, Academy of Management Review 32,
794–816.
Barnett, M. L., and R. M. Salomon, 2012, Does it pay to be really good? Addressing the
shape of the relationship between social and financial performance, Strategic
Management Journal 33, 1304–1320.
Brammer, S., and A. Millington, 2008, Does it pay to be different? An analysis of the
relationship between corporate social and financial performance, Strategic Manage-
ment Journal 29, 1325–1343.
Bridges, S., W. Marcum, and J. K. Harrison, 2003, The relation between employee
perceptions of stakeholder balance and corporate financial performance, SAM
Advanced Management Journal 68, 50–57.
Buchanan, B., C. X. Cao, and C. Chen, 2018, Corporate social responsibility, firm
value, and influential institutional ownership, Journal of Corporate Finance 52,
73–95.
Button, S. B., 2001, Organizational efforts to affirm sexual diversity: a cross-level
examination, Journal of Applied Psychology 86, 17–28.
Cahan, S. F., C. Chen, C. Li, and N. H. Nguyen, 2015, Corporate social responsibility
and media coverage, Journal of Banking and Finance 59, 409–422.
Cahan, S. F., C. Chen, and L. Chen, 2020, Do local social norms affect investors’
involvement in social activism? Revisiting the case of US institutional investors,
Accounting and Finance, https://doi.org/10.1111/acfi.12649.
Capon, N., J. U. Farley, and S. Hoenig, 1990, Determinants of financial performance: a
meta-analysis, Management Science 36, 1143–1159.
Cascio, W. F., 1991, Costing Human Resources: The Financial Impact of Behavior in
Organizations, 3rd edn (PWS-Kent Publishing, Boston, MA).
Chen, Y., and J. Jermias, 2014, Business strategy, executive compensation and firm
performance, Accounting and Finance 54, 113–134.
Chen, J., W. S. Leung, and K. P. Evans, 2016, Are employee-friendly workplaces
conducive to innovation?, Journal of Corporate Finance 40, 61–79.
Chintrakarn, P., S. Treepongkaruna, P. Jiraporn, and S. M. Lee, 2020, Do LGBT-
supportive corporate policies improve credit ratings? An instrumental variable
analysis, Journal of Business Ethics 162, 31–45.
Day, N. E., and P. Greene, 2008, A case for sexual orientation diversity management in
small and large organizations, Human Resource Management 47, 637–654.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 239
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Day, N. E., and P. Schoenrade, 1997, Staying in the closet vs. coming out: relationships
between communication about sexual orientation and work attitudes, Personnel
Psychology 50, 147–163.
Day, N. E., and P. Schoenrade, 2000, The relationship among reported disclosure of
sexual orientation, anti-discrimination policies, top management support and work
attitudes of gay and lesbian employees, Personnel Review 29, 346–363.
Dimson, E., O. Karakas, and X. Li, 2015, Active ownership, The Review of Financial
Studies 28, 3225–3268.
Donaldson, T., and L. E. Preston, 1995, The stakeholder theory of the corporation:
concepts, evidence, and implications, Academy of Management Review 20, 65–91.
Eccles, R. G., I. Ioannou, and G. Serafein, 2014, The impact of corporate sustainability
on organizational processes and performance, Management Science 60, 2835–2857.
Edmans, A., 2011, Does the stock market fully value intangibles? Employee satisfaction
and equity prices, Journal of Financial Economics 101, 621–640.
Edmans, A., 2012, The link between job satisfaction and firm value, with implications
for corporate social responsibility, Academy of Management Perspectives 26, 1–19.
Everly, B. A., and J. L. Schwarz, 2015, Predictors of the adoption of LGBT-friendly HR
policies, Human Resource Management 54, 367–384.
Faleye, O., and E. A. Trahan, 2011, Labor-friendly corporate practices: is what is good
for employees good for shareholders?, Journal of Business Ethics 101, 1–27.
Fauver, L., M. B. McDonald, and A. G. Taboada, 2018, Does it pay to treat employees
well? International evidence on the value of employee-friendly culture, Journal of
Corporate Finance 50, 84–108.
Fernando, S., and S. Lawrence, 2014, A theoretical framework for CSR practices:
integrating legitimacy theory, stakeholder theory and institutional theory, Journal of
Theoretical Accounting Research 10, 149–178.
Filbeck, G., and D. Preece, 2003, Fortune’s Best 100 Companies to Work for in
America: do they work for shareholders?, Journal of Business Finance and Accounting
30, 771–797.
Flammer, C., 2015, Does corporate social responsibility lead to superior financial
performance? A regression discontinuity approach, Management Science 61,
2549–2568.
Flammer, C., and J. Luo, 2017, Corporate social responsibility as an employee
governance tool: evidence from a quasi-experiment, Strategic Management Journal 38,
163–183.
Freeman, R. E., 1984, Strategic Management: A Stakeholder Approach (Pitman, Boston,
MA).
Friedman, M., 1962, Capitalism and Freedom (University of Chicago Press, Chicago,
IL).
Friedman, M., 1970, The social responsibility of business is to increase its profits, New
York Times Magazine (13 September), 122–126.
Frijns, B., O. Dodd, and H. Cimerova, 2016, The impact of cultural diversity in
corporate boards on firm performance, Journal of Corporate Finance 41, 521–541.
Gelade, G. A., and M. Ivery, 2003, The impact of human resource management and
work climate on organizational performance, Personnel Psychology 56, 383–404.
Griffin, J. J., and J. F. Mahon, 1997, The corporate social performance and corporate
financial performance debate: twenty-five years of incomparable research, Business
and Society 36, 5–31.
Griffith, K. H., and M. R. Hebl, 2002, The disclosure dilemma for gay men and lesbians:
‘coming out’ at work, Journal of Applied Psychology 87, 1191–1199.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
240 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Hasan, I., N. Kobeissi, L. Liu, and H. Wang, 2018, Corporate social responsibility and
firm financial performance: the mediating role of productivity, Journal of Business
Ethics 149, 671–688.
Henderson, P. D., 2001, Misguided Virtue: False Notions of Corporate Social
Responsibility (Institute of Economic Affairs, London).
Hilary, G., and K. W. Hui, 2009, Does religion matter in corporate decision making in
America?, Journal of Financial Economics 93, 455–473.
Huang, D. Z. X., 2021. Environmental, social and governance (ESG) activity and firm
performance: a review and consolidation. Accounting and Finance 61, 335–360.
Huffman, A. H., K. M. Watrous-Rodriguez, and E. B. King, 2008, Supporting a diverse
workforce: what type of support is most meaningful for lesbian and gay employees?,
Human Resource Management 47, 237–253.
Huselid, M. A., 1995, The impact of human resource management practices on
turnover, productivity, and corporate financial performance, Academy of Manage-
ment Journal 38, 635–672.
Jackson, S. E., A. Joshi, and N. L. Erhardt, 2003, Recent research on team and
organizational diversity: SWOT analysis and implications, Journal of Management 29,
801–830.
Jensen, M. C., 2002, Value maximization, stakeholder theory, and the corporate
objective function, Business Ethics Quarterly 12, 235–256.
Johnston, D., and M. A. Malina, 2008, Managing sexual orientation diversity: the
impact on firm value, Group and Organization Management 33, 602–625.
Jones, T. M., 1995, Instrumental stakeholder theory: a synthesis of ethics and
economics, Academy of Management Review 20, 404–437.
Kaplan, D. M., 2006, Can diversity training discriminate? Backlash to lesbian, gay, and
bisexual diversity initiatives, Employee Responsibilities and Rights Journal 18, 61–72.
Kitzmueller, M., and J. Shimshack, 2012, Economic perspectives on corporate social
responsibility, Journal of Economic Literature 50, 51–84.
Kochan, T., K. Bezrukova, R. Ely et al., 2003, The effects of diversity on business
performance: report of the diversity research network, Human Resource Management
42, 3–21.
Kru¨ger, P., 2015, Corporate goodness and shareholder wealth, Journal of Financial
Economics 115, 304–329.
Laplume, A. O., K. Sonpar, and R. A. Litz, 2008, Stakeholder theory: reviewing a
theory that moves us, Journal of Management 34, 1152–1189.
Li, F., and V. Nagar, 2013, Diversity and performance, Management Science 59,
529–544.
McWilliams, A., and D. S. Siegel, 2000, Corporate social responsibility and financial
performance: correlation or misspecification?, Strategic Management Journal 21,
603–608.
Metcalf, H., and H. Rolfe, 2011, Barriers to Employees in Developing Lesbian, LGBT,
Bisexual and Transgender-Friendly Work Places (National Institute of Economic and
Social Research, London).
Miller, S. R., L. Eden, and D. Li, 2020, CSR reputation and firm performance: a
dynamic approach, Journal of Business Ethics 163, 619–636.
Orlitzky, M., 2001, Does firm size confound the relationship between corporate social
performance and financial performance?, Journal of Business Ethics 33, 167–180.
Orlitzky, M., F. L. Schmidt, and S. L. Rynes, 2003, Corporate social and financial
performance: a meta-analysis, Organization Studies 24, 403–441.
Peloza, J., 2009, The challenge measuring financial impacts from investments in
corporate social performance, Journal of Management 35, 1518–1541.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
V. Fatmy et al./Accounting & Finance 62 (2022) 213–242 241
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Pichler, S., J. L. Blazovich, K. A. Cook, J. M. Huston, and W. R. Strawser, 2018, Do
LGBT-supportive corporate policies enhance firm performance?, Human Resource
Management 57, 263–278.
Ragins, B. R., and J. M. Cornwell, 2001, Pink triangles: antecedents and consequences
of perceived workplace discrimination against gay and lesbian employees, Journal of
Applied Psychology 86, 1244–1261.
Ragins, B. R., R. Singh, and J. M. Cornwell, 2007, Making the invisible visible: fear and
disclosure of sexual orientation at work, Journal of Applied Psychology 92, 1103–1118.
Richard, O. C., 2000, Racial diversity, business strategy, and firm performance: a
resource-based view, Academy of Management Journal 43, 164–177.
Roumpi, D., P. Giannakis, and J. E. Delery, 2020, Adoption of LGBT-friendly
practices: the effect of institutional pressures and strategic choice, Human Resource
Management Journal 30, 604–623.
Servaes, H., and A. Tamayo, 2013, The impact of corporate social responsibility on firm
value: the role of customer awareness, Management Science 59, 1045–1061.
Shan, L., S. Fu, and L. Zheng, 2017, Corporate sexual equality and firm performance,
Strategic Management Journal 38, 1812–1826.
Snider, J., R. P. Hill, and D. Martin, 2003, Corporate social responsibility in the 21st
century: a view from the world’s most successful firms, Journal of Business Ethics 48,
175–187.
Sundaram, A. K., and A. C. Inkpen, 2004, The corporate objective revisited,
Organization Science 15, 350–363.
Surroca, J., and J. A. Tribo´, 2008, Managerial entrenchment and corporate social
performance, Journal of Business Finance and Accounting 35, 748–789.
Surroca, J., J. A. Tribo´, and S. Waddock, 2010, Corporate responsibility and financial
performance: the role of intangible resources,StrategicManagement Journal 31, 463–490.
Trau, R. N. C., 2015, The impact of discriminatory climate perceptions on the
composition of intra-organizational developmental networks, psychosocial support,
and job and career attitudes of employees with an invisible stigma, Human Resource
Management 54, 345–356.
Unsal, O., and J. Brodmann, 2020, The impact of employee relations on the reputation
of the board of directors and CEO, The Quarterly Review of Economics and Finance
78, 372–388.
van Beurden, P., and T. Go¨ssling, 2008, The worth of values – a literature review on the
relation between corporate social and financial performance, Journal of Business
Ethics 82, 407–424.
Waddock, S. A., and S. B. Graves, 1997, The corporate social performance-financial
performance link, Strategic Management Journal 18, 303–319.
Wang, P., and J. L. Schwarz, 2010, Stock price reactions to GLBT non-discrimination
policies, Human Resource Management 49, 195–216.
Weinzimmer, L. G., and C. A. Esken, 2016, Risky business: taking a stand on social
issues, Business Horizons 59, 331–337.
Wettstein, F., and D. Baur, 2016, ‘Why should we care about marriage equality?’:
political advocacy as a part of corporate responsibility, Journal of Business Ethics 138,
199–213.
Whitener, E. M., 2001, Do ‘high commitment’ human resource practices affect employee
commitment? A cross-level analysis using hierarchical linear modelling, Journal of
Management 27, 515–535.
Wright, P., S. P. Ferris, J. S. Hiller, and M. Kroll, 1995, Competitiveness through the
management of diversity: effects on stock price valuation, Academy of Management
Journal 38, 272–287.
© 2021 The Authors. Accounting & Finance published by John Wiley & Sons Australia,
Ltd on behalf of Accounting and Finance Association of Australia and New Zealand
242 V. Fatmy et al./Accounting & Finance 62 (2022) 213–242
 1467629x, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/acfi.12787 by Duodecim Medical Publications Ltd, Wiley Online Library on [21/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License