Journal of Corporate Finance 62 (2020) 101588

Contents lists available at ScienceDirect

Journal of Corporate Finance

journal homepage: www.elsevier.com/locate/jcorpfin

Corporate social responsibility, product market perception, and

T
firm value
Katsiaryna Salavei Bardosa, Mine Ertugrulb, , Lucia Silva Gaob
⁎

a
Fairfield University, 1073 North Benson Road Fairfield, CT 06824, United States of America
b
University of Massachusetts Boston, 100 Morrissey Blvd, Boston, MA 02125, United States of America

ARTICLE INFO ABSTRACT

Keywords: We examine whether corporate social responsibility (CSR) is used to signal product quality and
Corporate social responsibility whether CSR affects firm value through its positive effect on product market perception. Using a
Product market perception proprietary database, we find that visible CSR, such as environmental and community involve-
Brand value ment, positively impacts product market perception, particularly for standardized goods and in
Firm value
competitive industries, and that this impact is more pronounced for product quality attributes.
Furthermore, we find that CSR indirectly increases firm value through an improvement in pro-
duct market perception. We conclude that product market perception is a channel through which
CSR creates firm value.

1. Introduction

“Corporate social responsibility (CSR) occurs when firms engage in activity that appears to advance social agenda beyond that
which is required by law.” (Siegel and Vitaliano, 2007). The importance of CSR has been growing in the last few years. In January of
2018, Larry Fink, the CEO of BlackRock, called for the corporate CEOs to think not just about profits, but also about making a
“positive contribution to society”.1 In August of 2019 more than 180 of CEOs pledged that their firms' purpose was no longer to serve
their owners alone, but customers, employees, suppliers and communities, too.2
Harjoto and Jo (2011) summarize the existing theories as to why firms undertake CSR and argue that one of the ways firms use
CSR is to signal product quality.3 Such signals deal with adverse selection arising from information asymmetry about product quality
(Kirmani and Rao, 2000). As a result, while CSR is costly, it has strategic implications and is a source of competitive advantage for
companies across different industries (Baron, 2001; McWilliams et al., 2006; Porter and Kramer, 2006). However, the literature on
the empirical relationship between CSR and firm value is inconclusive, with many of the studies showing a positive impact of CSR on
firm value, but some providing evidence in the opposite direction.4 This relationship may be unclear because of the lack of un-
derstanding about the mechanisms through which CSR affects firm value (Servaes and Tamayo, 2013). Several studies argue that

Corresponding author.
⁎

E-mail addresses: kbardos@fairfield.edu (K.S. Bardos), mine.ertugrul@umb.edu (M. Ertugrul), lucia.silva-gao@umb.edu (L.S. Gao).
1
Larry Fink's Annual Letter to CEOs: A Sense of Purpose, https://www.blackrock.com/hk/en/insights/larry-fink-ceo-letter.
2
Business Roundtable. Statement on the Purpose of Corporation, https://opportunity.businessroundtable.org/ourcommitment/
3
Other theories of CSR argue that CSR is a result of the principal agent problem and that top management overinvests in CSR for their own
interests; that investment in CSR is meant to reduce the probability of shareholder turnover; or that CSR aligns the interests of investing and non-
investing stakeholders.
4
See, for example, Garcia-Castro et al. (2010), Margolis and Walsh (2003), Margolis et al. (2007), McWilliams and Siegel (2000), Orlitzky (2001),
Orlitzky et al. (2003), Van Beurden and Gössling, 2008.

https://doi.org/10.1016/j.jcorpfin.2020.101588
Received 2 November 2018; Received in revised form 13 November 2019; Accepted 1 February 2020
Available online 07 February 2020
0929-1199/ © 2020 Elsevier B.V. All rights reserved.
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

there is an indirect link between CSR and firm value (e.g. Servaes and Tamayo, 2013; Saeidi et al., 2015; Galbreath and Shum, 2012).
However, the channels through which CSR creates value are still not well understood.
This paper investigates if CSR affects firm value through improving perceived product quality and serving as a product differ-
entiation strategy. More specifically, we investigate whether CSR activities, especially those that are visible to customers, such as
environmental and community CSR, have an effect on product market perception, and indirectly on firm value.
A survey by Accenture and United Nations Global Compact finds that 72% of the CEOs consider “brand, trust, and reputation”
(2010:14) as the main reasons for undertaking CSR (Flammer, 2013). A survey of executives and investors conducted by the
Economist Intelligence Unit in 2005 found that most of them (61%) believe that brand enhancement is the most important business
benefit of CSR.5 After reviewing the literature, Kitzmueller and Shimshack (2012) conclude that firms use CSR primarily to differ-
entiate their product and signal its quality. Furthermore, the resource-based view of the firm suggests that companies may engage in
CSR to enhance their brand, reputation, and trust (Barney, 1991; Porter, 1991; Porter and Kramer, 2006, 2011).
The literature on the relationship between CSR and firm value usually refers to the stakeholder theory, which predicts that CSR
positively impacts shareholders' wealth because focusing on the interests of other stakeholders increases their willingness to support
firms' operations. CSR may improve customer perception and satisfaction, and in this way contribute to the enhancement of product
market perception. Although the customer channel is not the only possible channel that can explain the relationship between CSR and
firm value, customers' perception and behavior clearly affect a company's financial performance and value. We focus on the CSR
activities that are most visible to customers, more specifically environmental and community CSR, consistent with findings from
previous studies.6
Using a large sample of companies across different industries for the period 2001–2014, we start by providing evidence of the
relationship between CSR and product market perception. Previous studies on this relationship are based on small-scale surveys,
small samples, or have a limited focus.7 We use a large sample based on a proprietary database of customer brand evaluation. The
measures we use rely on a customer survey-based approach and, therefore, reflect the product market perception, in contrast with
models based on financial measures or expert evaluation, which do not reflect customers' perception.
We find that CSR positively affects product market perception. This result is economically meaningful: one standard deviation
increase in the CSR measure increases the product market perception measure by 10.5%. The impact is significant for both the
community and environmental components of CSR. When breaking CSR into strengths and concerns, we find that both community
and environmental strengths positively impact product market perception, but the negative impact of concerns is not significant. This
result suggests that the strengths components of CSR are more visible. We find that there is no association between most other
components of CSR, namely, employee friendliness, diversity, corporate governance, and product market perception. In addition, the
positive association between community and environmental CSR is most pronounced for firms with standardized rather than dif-
ferentiated goods, and for companies in competitive industries. We also find that our results are most pronounced for a more refined
measure of perceived product quality.
The results that show a positive association between community and environmental CSR and product market perception are
subject to endogeneity concerns. Our results may suffer from reverse causality: it is possible that firms with strong product market
perception can afford to spend more on CSR. Another potential issue that could impact our results is the omitted variables bias. We
address these issues in several ways. First, we do an instrumental variable analysis using per capita CO2 emissions from fossil fuel
combustion and the percentage of population that volunteer for non-profit and community organizations in the state where the firm is
headquartered as instruments. Second, we do a quasi-natural experiment to examine the effect of BP Oil Spill on product market
perception of firms and find that the impact of environmental and community related CSR activities on product perception is stronger
after the spill for energy firms. This result suggests that these activities become more important after public relations shock to the
firms in the energy industry. Overall, these analyses suggest that our results are robust to endogeneity concerns.
After showing that CSR is associated with a favorable product market perception, we analyze the indirect link between CSR and
firm value, as measured by the Tobin's Q and profit margin. We find that the product market perception is significantly positively
related to firm value and profit margins. One standard deviation change in product market perception increases firm value by 5.8%.
Mediation analysis suggests that partial mediation occurs when product market perception is included in the Tobin's Q regression
together with CSR. These results confirm our prediction that product market perception is a channel through which CSR creates firm
value.
Our paper contributes to a growing body of finance literature on CSR (e.g., Ferrell et al., 2016; Kruger, 2015; Mishra, 2017;
Erhemjamts et al., 2013; Deng et al., 2013; Fatemi et al., 2015; Renneboog et al., 2008; Liang and Renneboog, 2017; Lins et al., 2017;
Giuli and Kostovetsky, 2014; Boone and Uysal, 2018; Adhikari, 2016). It also contributes to a relatively new examination of product

5
Economist Intelligence Unit, 2005. The importance of corporate responsibility.
6
Kruger (2015) finds that the economic magnitude of the short-term market reaction is most pronounced for environmental and community
related news announcements. Fisman et al. (2006) state that community CSR is the most visible aspect of CSR. Flammer (2013) finds that there is a
positive stock market reaction to the announcements of eco-friendly initiatives, and a negative stock market reaction to the announcements of eco-
harmful behavior. Flammer et al. (2017) find that the integration of CSR criteria in executive compensation is more pronounced with regard to
environmental and community CSR (related to “dependent” stakeholders).
7
Lai et al. (2010) is based on a survey among purchasing managers of Taiwanese manufacturing and service companies. Castaldo et al. (2009)
surveyed Italian clients of retail chains offering Fair Trade products. Sen and Bhattacharya (2001) use a survey of 277 MBA students. Hsu (2012) is
based on a survey conducted on policyholders of insurance companies in Taiwan. Hur et al. (2014) is based on a sample of 867 consumers surveyed
in South Korea. Melo and Galan (2011) uses data from Interbrand for 47 companies.

2
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

market perception (Larkin, 2013; Frieder and Subrahmanyam, 2005). This paper is the first to show the heterogeneous effect of CSR
on product market perception for standardized versus differentiated goods industries. This paper is also the first to examine product
market perception as a channel through which CSR affects firm value. Thus, we contribute to the literature that suggests that CSR has
an indirect effect on firm value (Servaes and Tamayo, 2013; Saeidi et al., 2015; Luo and Bhattacharya, 2006). We also contribute to
the large body of literature that finds a positive effect of CSR on firm value, supporting the stakeholder view of the firm (e.g., Orlitzky
et al., 2003; Van Beurden and Gössling, 2008; McWilliams et al., 2006).
Our paper is different from previous studies on CSR and customer perception in several ways. We establish a direct relationship
between CSR and product market perception. Our results suggest that the impact of CSR on product market perception is more
significant for standardized goods and in competitive industries. Servaes and Tamayo (2013) use advertising expenditures as a
measure of customer awareness. They argue that advertising creates awareness about CSR among customers and can be a product
differentiating strategy that enhances firm value. However, they do not test the direct impact of CSR on product market perception.
Furthermore, advertising is not a good proxy for brand value and consumer loyalty (Larkin, 2013, page 235), since it could facilitate
competition rather than create barriers to entry, and may be just one of the many tools used in strategic brand management. Our
measure, on the other hand, captures the outcome of strategic brand management. Our paper extends the results in Melo and Galan
(2011) and Torres et al. (2012), who find that CSR has a positive effect on product market perception. However, both of these papers
use a much smaller sample and measures of product market perception that do not reflect customers' brand perception.8 We also
contribute to the literature by controlling for endogeneity and examining the effect of CSR in standardized versus differentiated
product, as well as competitive industries.
The rest of the paper is organized as follows: Section 2 reviews previous literature; Section 3 describes the data; Section 4 presents
and discusses the results of the analysis of the relationship between CSR and product market perception, and the indirect link between
CSR and firm value. Section 5 concludes.

2. Related literature and hypotheses development

2.1. CSR and product market perception

CSR activities may have a positive effect on product market perception through the improvement of a company's image and
reputation (Hur et al., 2014; Jones, 2005; Porter and Kramer, 2006). Baron (2001) coins the term “strategic CSR” and states that
companies compete for socially responsible customers. Siegel and Vitaliano (2007) claim that CSR “is likely to be integrated into the
company's business-level product differentiation strategies.” As a result, firms may use CSR to signal their product quality (Harjoto
and Jo, 2011).
Previous research suggests that customers take into consideration firms' CSR activities when making purchase decisions, and are
more likely to purchase goods from more socially responsible firms, or even willing to pay a higher price (e.g., Sen and Bhattacharya,
2001; Bhattacharya and Sen, 2004; Penn Schoen Berland, 2010, Kitzmueller and Shimshack, 2012). Kitzmueller and Shimshack
(2012) state that “firms use CSR to differentiate and advertise their product or to build brand loyalty” and “CSR is meant to transmit a
positive signal about firm quality and type.” Such signaling resolves adverse selection situations that arise when buyers are unsure
about the true quality of the seller's product and constitutes a sales-independent signal (Kirmani and Rao, 2000).
Fisman et al. (2006) develop a theoretical model in which CSR signals product quality. They show that firms that care about
externalities from product quality engage in CSR to signal product quality. Consequently, CSR activities that are visible to consumers
are useful in signaling the firm's trustworthiness in providing quality products.
The aforementioned literature suggests that CSR activities are generally perceived to increase the product market perception of a
company and its products. Therefore, we expect a positive relationship between the level of CSR engagement and product market
perception. We state the following hypothesis:
H1. Product market perception increases with the level of CSR engagement.

2.1.1. Market competition

Fisman et al. (2006) propose that CSR is a source of product differentiation in competitive industries. They argue that CSR signals
the trustworthiness of the firm in providing (unobservable) quality and may be a way for firms to vertically differentiate themselves
in a market where quality is difficult to observe. Therefore, they predict more CSR engagement in competitive industries because CSR
serves as a source of product differentiation. However, Bagnoli and Watts (2003) predict more CSR engagement in a less competitive
environment. They model CSR as a private provision of public goods at levels that vary inversely with the degree of competitiveness
in the private goods markets. Therefore, there is no consensus among the theoretical predictions of the relationship between CSR and
competition.

8
Melo and Galan (2011) analyze 48 companies over the 2001–2003 time period. Therefore, they use a very limited panel data. Their brand
measure is based on the “Most Valuable Brands Report” provided by the consultancy firm Interbrand. Interbrand analysts evaluate the brand in
terms of financial performance, role of the brand in purchase decisions, and the competitive strength of the brand. The coefficient estimate on CSR in
Melo and Galan's analysis is not significant in regressions with one-year lagged brand measures for all seven qualitative areas of MSCI ESG ratings.
Torres et al. (2012) analyze 57 international companies and also use brand measures based on Interbrand provided data.

3
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Harjoto and Jo (2011) find empirical support for the hypothesis that firms in more competitive markets, as measured by higher
advertising ratios, are more likely to engage in CSR. They also find a positive association between advertising expenditures, CSR and
financial performance. Given these empirical findings and the arguments introduced by Fisman et al. (2006), we expect the impact of
CSR on product market perception to be more pronounced in competitive industries. Thus, we offer the following hypothesis:
H1A. The positive association between product market perception and CSR is more pronounced in competitive industries.

2.1.2. Standardized vs. differentiated goods

It has also been shown that the more unique the product the more customers value it (Tian et al., 2001). Differentiated goods are
harder to replace because they provide a unique service or product, and therefore have higher switching costs (Giannetti et al., 2011).
Albuquerque et al. (2017) predict that CSR is associated with lower firm-level systematic risk for differentiated goods. They assume
that greater product differentiation is a proxy for lower elasticity of substitution. We predict that CSR is associated with more positive
product market perception in standardized rather than differentiated product industries, because these industries already have high
product market perception and, therefore, the impact of CSR on product market perception should be lower. Furthermore, if CSR is a
source of product differentiation and a signal of product quality, it will be more important in industries in which the products are
more standardized, and thus the effect of CSR on product market perception will be more pronounced in standardized goods in-
dustries. Therefore, we test the following hypothesis:
H1B. The positive association between product market perception and CSR is more pronounced in standardized goods industries.

2.2. Product market perception and firm value

The resource-based view (RBV) offers one perspective to explain the value enhancement of product market perception. According
to the RBV, value is derived from corporate reputation as an important strategic asset that differentiates a company from its com-
petitors and is difficult for competing firms to replicate (Fombrun and Shanley, 1990; Roberts and Dowling, 2002). Long-term
reputation can be maintained and improved by increasing customer satisfaction (Anderson and Sullivan, 1993; Galbreath and Shum,
2012). Galbreath and Shum (2012) further contend that the relation between customer satisfaction and firm performance is entirely
mediated by reputation and thus corporate reputation seems to be the driver of value from customer satisfaction. Furthermore,
research on the relationship between reputation and firm performance shows financial benefits from good reputation. For example,
good reputation is associated with lower firm risk (Helm, 2007) and higher sales and return on assets (Kotha et al., 2001; Roberts and
Dowling, 2002). Therefore, we predict a positive association between product market perception and firm value.
Empirical studies provide additional support to the association between product market perception and value. Using the same
database as we use in this paper, Larkin (2013) examines the implications of brand perception for cash flow stability and financial
policy. She finds that positive brand perception lowers cash flow volatility, improves credit ratings, increases leverage, and lowers
cash holdings. However, Larkin (2013) suggests that it is not clear what effect brand perception will have on firm value. On one hand
favorable brand perception reduces cash flow volatility, which should improve firm value, but on the other hand it is also associated
with increased leverage. Also using this database, Mizik and Jacobson (2008) examine which brand perception metrics (differ-
entiation, relevance, esteem, knowledge, and energy) explain stock returns. They find that the relationship between brand perception
and stock returns is significant for brand relevance and energy, but not for esteem and knowledge. Differentiation does not appear to
have incremental information content.

2.3. CSR and firm value

Two existing theories have opposite predictions of the relationship between CSR and firm value: the stakeholder value max-
imization view and the shareholder expense view (Gregory and Whittaker, 2013).9 The stakeholder theory posits that CSR has a
positive effect on shareholder wealth because focusing on the interests of other stakeholders increases their willingness to support
firm's operations. This argument is in line with the contract theory and theory of the firm, which views the firm as a nexus of contracts
between shareholders and other stakeholders that supply critical resources. These contracts can be explicit or implicit, and firms can
default on the implicit contracts. The value of these contracts depends on stakeholders' expectations of the firm honoring its com-
mitments. CSR initiatives contribute to increasing the firm's reputation for keeping its commitments, and therefore increase the
incentives of stakeholders to contribute with resources and effort to the firm (Freeman, 1984; Jensen, 2001; Freeman and McVea,
2001; Freeman et al., 2004). Therefore, CSR improves financial performance by improving the relationships of a firm with its
stakeholder groups.
The shareholder expense view suggests that CSR is undertaken at the expense of shareholders and, therefore, lowers firm value
(Friedman, 1970; Friedman, 1998; Cronqvist et al., 2009; Pagano and Volpin, 2005). Friedman suggests that the mere existence of
CSR is a manifestation of agency problems. The agency theory perspective implies that CSR expenditures are a misuse of funds that
should be used for projects that add value to shareholders, and that CSR expenditures are an executive perk (McWilliams et al., 2006).
In addition, the empirical research on the relationship between CSR and financial performance is not clear, with most of the

9
The “Freeman versus Friedman” proposition (Gregory and Whittaker, 2013).

4
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

studies showing a positive relationship, but some finding a negative relationship, or no relationship (e.g., Garcia-Castro et al., 2010;
Margolis and Walsh, 2003; Margolis et al., 2007; McWilliams and Siegel, 2000; Orlitzky, 2001; Orlitzky et al., 2003; Van Beurden and
Gössling, 2008; Bhandari and Javakhadze, 2017). For example, Kruger (2015) examines short-term market reaction to CSR news
announcements and finds strong negative reaction to negative events and weak negative reaction to positive events. He also finds that
improving CSR can be value enhancing when CSR news are aimed at offsetting prior corporate social irresponsibility, and the reaction
is more pronounced for CSR news that contain strong economic and legal information. Flammer (2013) finds a positive market
reaction to positive environmental CSR announcements and a negative reaction to negative announcements. Margolis et al. (2007)
find a modest positive average correlation between CSR and financial performance.
Despite the conflicting empirical results found in the literature, a vast body of research provides arguments for a positive impact
of CSR on firm value. Barnett (2007) argues that the impact of CSR on firm value depends on the ability of CSR to influence
stakeholders in the firm. Firms are able to charge premium prices because of the improved relation between the firm and its sta-
keholders.
Servaes and Tamayo (2013) examine the role of customer awareness on the impact of CSR on firm value. They differentiate
between signaling and consumer awareness arguments and suggest that a necessary condition for CSR to influence firm value is
consumer's awareness of CSR. They conjecture that advertising reduces information gap between the firm and its customers, which
makes it more likely that customers will find out about CSR and reward the firm for it. Using advertising expenditures as a proxy for
consumer awareness, they find a positive association between CSR and firm value only for firms with high levels of advertising
expenditures.
The aforementioned research suggests that CSR may directly impact firm value, but CSR also improves customer satisfaction,
reputation, and product market perception, which in turn have a positive effect on firm financial performance. Therefore, CSR may
impact firm value either directly or indirectly. We hypothesize that product market perception is a mechanism of value creation
through CSR, and has a mediating role on the relationship between CSR and firm performance. Therefore, we state the following
hypothesis:
H2. CSR increases firm value through the increase in product market perception.

3. Data

3.1. Product market perception

Brand Asset Valuator is a proprietary brand assessment model developed by BAV Consulting, a subsidiary of Young & Rubicam.
BAV surveys more than 16,000 US households to evaluate brands on a wide range of attributes. BAV Consulting conducted pilot
surveys in 1993 and 1997 and has been conducting the survey annually since 2001. We use the following attributes measured in the
survey to construct our Product Market Perception measure: 1) Relevance, 2) Knowledge, 3) Distinctive, 4) Unique, 5) Dynamic, 6)
Innovative, 7) Leader, 8) Reliable, 9) High-quality, 10) Trusthworthy.10 Since these measures are not all measured in the same scale
(for example Relevance is measured in a 1–7 scale, while Unique is the percentage of people surveyed that responded “yes”), we first
compute z-scores for each of these items across all brands and take their average.
The BAV questionnaire is conducted at the brand level. Thus, we manually link the brand to the companies. We follow Larkin
(2013) and identify the brand that is closest to the corporation's name and use its brand asset score. The details of this procedure
based on the type of brands is provided in the Appendix.
After we link the brands to companies, we merge this dataset with Compustat and MSCI ESG Research database (formerly known
as KLD). The resulting database has 2505 firm-year observations for 364 unique firms. In our sample, we use BAV surveys for 1997,
and 2001 to 2014, and MSCI data for 1996, and 2000 to 2013.11

3.2. Corporate social responsibility

We use MSCI ESG Research database (formerly known as KLD Research and Analytics Database) to construct measures of social
and environmental performance. MSCI provides social responsibility research and indexes for institutional investors. It gathers data
about companies from a variety of sources such as company filings, general media sources, annual questionnaires sent to companies'
investor relations offices, academic publications and government data. After information is collected, an analyst from a sector-specific
research team evaluates and rates the firm based on screens called “strengths” and “concerns” in seven major areas: Community,
corporate governance, diversity, employee relations, environment, human rights, product safety and quality. In this paper, we are
interested in the ratings that are visible to customers. Thus, we focus on the screens for community and environment. The list of the

10
These attributes are also used by BAV Consulting to create their Brand Asset measures. The only difference is that we substituted “Personal
Regard” with “Trustworthy” since the former measure was not available to us. For a more detailed description of these measures, please see Mizik
and Jacobson (2008). We choose not to use BAV's Brand Asset composite measure since BAV does not report how they combined these attributes
into one measure. Instead, we average the standardized components of the measure to create our measure. The results using BAV's composite
measure are similar to the ones reported in this paper.
11
Since our MSCI ESG data starts in 1995, we exclude the BAV survey results for 1993 from our analysis.

5
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

screens we use in our analyses is as shown below.

• Community Strengths: Charitable giving, innovative giving, support for housing, support for education, non-US charitable giving,
volunteer programs, community engagement, other community strength.
• Community Concerns: Investment controversies, negative economic impact, tax disputes, other community concern.
• Environmental Strengths: Beneficial products and services, pollution prevention, recycling, clean energy, communications, and
other strength, management systems strength, water stress, biodiversity and land use, raw material sourcing, natural resource use,
green buildings, renewable energy, waste management, energy efficiency, product carbon footprint, insuring climate change risk.
• Environmental Concerns: Hazardous waste, regulatory problems, ozone depleting chemicals, substantial emissions, agricultural
chemicals, climate change, and other concerns, negative impact of products and services, and use and biodiversity, non‑carbon
releases, supply chain management.

MSCI refines its ratings every several years, thus changing the number of strengths and concerns in each rating category. For
example there are 6 environmental concern screens in 1997, but 7 in 1998. In order to make strengths or concerns comparable within
each category across years, we scale the strengths and concerns by dividing the number of strengths (concerns) for each firm- year
within each CSR category by the maximum possible number of strengths (concerns) in each category-year. Thus, our indices of
strengths and concerns range from 0 to 1.

3.3. Control variables

We include several firm-level controls in our analysis. These data was obtained from Compustat. We control for firm size (log of
total assets), market-to-book ratio (the ratio of market value of assets to total assets), leverage (the ratio of total debt to total assets),
return on assets (the ratio of operating income before depreciation to total assets), advertising expenses over sales, research and
development expenses over sales, and selling, general, and administrative expenses over sales. Following prior literature, advertising
expenses, research and development, and selling, general and administrative expenses are set equal to zero when missing. In addition,
we control for factors that could significantly affect product market perception. We control for corporate reputation with indicator
variables for whether the firm has faced class action lawsuits for that year, has restated its financial statements, and has cut its
dividend. We obtained the data to construct these variables from the Stanford Law School & Cornerstone Research database, Audit
Analytics, and Compustat, respectively. We also control for managerial ability, organizational capital, and boardroom reputation. For
managerial ability, we use the measure developed in Demerjian et al. (2012).12 Following Lev and Radhakrishnan (2005) and Eisfeldt
and Papanikolaou (2013), we measure organizational capital using capitalized SG&A expense. For boardroom reputation, we include
board size, percentage of independent directors, and the ratio of directors' equity based compensation to their total compensation. We
winsorize all Compustat variables (except log of total assets) at the top and bottom 1%. We also include Fama-French 48 industry and
year dummies in our regressions.

3.4. Summary statistics

Table 1 shows the summary statistics for the data. The firms in the sample are quite large, with a median total asset value of
around $5.9 billion. The median firm in the sample has a market- to-book ratio of 1.79, and is highly levered, with a debt ratio of
43.5%. At the median, sample firms are also quite profitable, with a ROA ratio of 15.3%.
The last column of Table 1 shows the correlation of MSCI ESG ratings and firm variables with the product market perception
measure. The Community/Environment CSR measure is positively correlated with product market perception at 13.5%. The product
market perception measure is also positively correlated with firm size, market-to-book, advertising expenses, operating profitability,
and managerial ability. It has a negative correlations with leverage, restatement, dividend cut, and organizational capital.
Table 2 compares the industry distribution of the sample firms with Compustat firms. The BAV sample is more heavily weighted
towards consumer durables and retail. This is expected, since product market perception is more important in business-to-consumer
industries, and less so in business-to-business industries. However, the overall sample includes a wide variety of companies from
different industries. Our sample is similar to that analyzed by Larkin (2013). Our sample excludes financial firms and utilities.

4. Results

4.1. CSR and product market perception

In Table 3, we examine the relation between community and environmental CSR and product market perception (Product Per-
ception for short) using the following regressions specification:
Product Perceptioni , t + 1 = + CSRi . t + Firm controlsi. t + Industry dummiesi . t + Year dummiest + i. t (1)

12
The measure is generated using Data Envelopment Analysis and measures managers' efficiency in generating revenues from a given set of
inputs. The measure is available for download at http://faculty.washington.edu/pdemerj/data.html.

6
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 1
Summary statistics.
Mean Median Std. Dev. Correlation with brand value

Product Perception 0.152 0.055 0.649 1.000

Community/Environmental CSR 0.059 0.000 0.268 0.135
Community CSR 0.056 0.000 0.206 0.137
Environmental CSR 0.070 0.000 0.216 0.149
Log(Assets) 8.671 8.682 1.747 0.256
MB ratio 2.183 1.793 1.374 0.163
Leverage 0.457 0.435 0.265 −0.014
ROA 0.157 0.153 0.095 0.093
Advertising/Sales 0.031 0.018 0.040 0.109
R&D/Sales 0.032 0.000 0.066 0.028
SG&A/Sales 0.257 0.236 0.161 0.043
Managerial Ability 0.636 0.700 0.315 0.173
Litigation 0.036 0.000 0.186 0.015
Restatement 0.054 0.000 0.227 −0.024
Dividend cut 0.154 0.000 0.361 −0.001
Organizational Capital 0.069 0.010 0.218 −0.087

This table reports summary statistics for the sample. The sample consist of 2505 firm-year observations. Product Perception an average of stan-
dardized values of the following: Relevance, Knowledge, Distinctive, Unique, Dynamic, Innovative, Leader, Reliable, High quality, and Trustworthy.
Community/Environmental CSR is the difference between community and environmental CSR strengths and concerns from MSCI ESG Research
database. Log (Assets) is the logarithm of total assets. MB ratio is the ratio of market value of assets to book value of assets. Leverage is the ratio of the
sum of current liability and long-term debt to total assets. ROA is the ratio of operating income to total assets. Advertising/Sales is the ratio of
advertising expense to sales. R&D/Sales is the ratio of research and development expense to sales. SG&A/Sales is the ratio of selling, general, and
administrative expenses to sales. Managerial Ability is the measure by Demerjian et al. (2012). Litigation, Restatement, and Dividend Cut are dummy
variables that equal one if the company faced class action lawsuits, restated its financials, and cut its dividend for that year, respectively. Orga-
nizational Capital is calculated following Eisfeldt and Papanikolaou (2013) and scaled by total assets. The last column of the table reports the
correlation of the variables with brand asset measure.

Table 2
Industry distribution.
Number of firms Market capitalization

BAV Compustat BAV Compustat

Business Equipment Computers, Software, and Electronic Equipment 15.53 16.26 24.31 13.13
Chemicals Chemicals and Allied Products 4.9 1.89 5.88 3.35
Consumer Durables Cars, TV's, Furniture, Household Appliances 3.79 2 1.19 2.59
Energy Oil, Gas, and Coal Extraction and Products 2.4 4.15 11.19 10.72
Health Healthcare, Medical Equipment, and Drugs 3.67 9.64 12.02 9.19
Manufacturing Machinery, Trucks, Planes, Office Furniture, Paper, Commercial Printing 11.66 7.58 8.13 6.37
Consumer Nondurables Food, Tobacco, Textiles, Apparel, Leather, Toys 17.37 4.01 9.42 6.52
Other Mines, Construction, Building Materials, 8.70 28.57 6.64 12.09
Shops Transportation, Hotels, Business Services, Entertainment 27.90 6.9 14.63 6.43
Wholesale, Retail, and Some Services
Telecommunications Telephone and Television Transmission 4.11 3 6.51 8.66

This table presents the Fama-French 12 industry distribution of Compustat firms and the sample firms. The distribution is calculated based on both
the number of firms and market capitalization of the firms in each industry category.

As stated in hypothesis H1, we predict a positive relation between CSR and product perception. The results in Column (1) show
that community and environmental CSR are significantly and positively correlated with product market perception, supporting our
prediction. The coefficient estimate of 0.392 indicates that one standard deviation increase in the CSR measure increases the product
perception measure by 10.5%. Columns (2) and (3) show the effect of community and environmental CSR on product market
perception separately. The results show that both of these CSR components are significantly and positively linked to product market
perception. One standard deviation increase in environmental CSR is associated with an 8.7% increase in product market perception,
while one standard deviation increase in community related CSR is associated with a 2.95% increase in product market perception.
Thus, the effect of environment related CSR on product market perception is stronger than that of community related CSR. The results
also show that product market perception is positively related to size and market-to-book ratio. In some specifications advertising
expenditures have positive and significant coefficient while leverage, ROA, R&D, litigation and dividend cut have negative and
significant coefficients. In untabulated results we replicated model (1) with 2-digit and 3-digit SIC code industry dummies and find
that our results are robust to the definition of industries. We also include the following board of directors related measures: board
size, percentage of independent directors, and the ratio of directors' equity based compensation to their total compensation in our
main specification (results untabulated). Our results remain robust to including these measures.

7
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 3
Product market perception and community and environmental CSR.
(1) (2) (3) (4) (5) (6)

Product Product Product Product Product Product

perceptiont+1 perceptiont+1 perceptiont+1 perceptiont+1 perceptiont+1 perceptiont+1

Community/Environmental 0.392***
CSR (0.000)
Community 0.143**
(0.045)
Environmental 0.401***
(0.001)
Community/Environmental 0.423***
CSR - Strength (0.000)
Community/Environmental -0.270
CSR - Concern (0.383)
Community - Strength 0.267***
(0.004)
Community - Concern 0.022
(0.836)
Environmental – Strength 0.419***
(0.001)
Environmental - Concern −0.341
(0.213)
Log (Assets) 0.154*** 0.166*** 0.156*** 0.148*** 0.153*** 0.154***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
MB ratio 0.093*** 0.095*** 0.093*** 0.092*** 0.094*** 0.093***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Leverage −0.092 −0.095 −0.093 −0.091 −0.095 −0.092
(0.192) (0.188) (0.195) (0.197) (0.184) (0.201)
ROA −0.265 −0.225 −0.247 −0.261 −0.235 −0.243
(0.360) (0.442) (0.398) (0.365) (0.417) (0.400)
Advertising/Sales 1.071 1.211 1.099 1.075 1.175 1.104
(0.181) (0.135) (0.165) (0.179) (0.147) (0.163)
R&D/Sales 0.094 0.078 0.122 0.085 0.045 0.118
(0.890) (0.909) (0.855) (0.899) (0.946) (0.860)
SG&A/Sales 0.195 0.239 0.187 0.200 0.244 0.190
(0.501) (0.413) (0.517) (0.489) (0.401) (0.508)
Managerial Ability 0.071 0.043 0.080 0.072 0.040 0.081
(0.238) (0.477) (0.175) (0.231) (0.517) (0.171)
Litigation −0.024 −0.013 −0.027 −0.024 −0.010 −0.028
(0.625) (0.782) (0.571) (0.618) (0.835) (0.562)
Restatement 0.030 0.036 0.038 0.029 0.033 0.037
(0.499) (0.426) (0.395) (0.512) (0.469) (0.398)
Dividend cut −0.049 −0.046 −0.051 −0.048 −0.042 −0.051
(0.217) (0.253) (0.198) (0.229) (0.292) (0.199)
Organizational Capital −0.041 −0.045 −0.042 −0.045 −0.048 −0.045
(0.565) (0.531) (0.556) (0.519) (0.502) (0.523)

Observations 2,505 2,505 2,505 2,505 2,505 2,505

R-squared 0.460 0.454 0.461 0.461 0.457 0.461
Year dummies Yes Yes Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes Yes Yes

This table reports the OLS regression results of Community and Environmental CSR measures on Product Perception. Product Perception is an average
of standardized values of the following: Relevance, Knowledge, Distinctive, Unique, Dynamic, Innovative, Leader, Reliable, High quality,
Trustworthy. Community/Environmental CSR is the difference between community and environmental CSR strengths and concerns from MSCI ESG
Research database. Details of the calculation of these measures are described in the text. Log (Assets) is the logarithm of total assets. MB ratio is the
ratio of market value of assets to book value of assets. Leverage is the ratio of the sum of current liability and long-term debt to total assets. ROA is the
ratio of operating income to total assets. Advertising/Sales is the ratio of advertising expense to sales. R&D/Sales is the ratio of research and
development expense to sales. SG&A/Sales is the ratio of selling, general, and administrative expenses to sales. Managerial Ability is the measure by
Demerjian, Lev, and McVay (2012). Litigation, Restatement, and Dividend Cut are dummy variables that equal one if the company faced class action
lawsuits, restated its financials, or cut its dividend for that year, respectively. Organizational Capital is calculated following Eisfeldt and Papanikolaou
(2013) and scaled by total assets. All models include Fama-French 48 industry and year dummies, and a constant. p-values based on robust standard
errors clustered at the firm level are in parentheses. *, **, and *** denote statistical significance at 10%, 5%, and 1% levels, respectively.

In Columns (4) to (6) of Table 3, we analyze the effect of CSR strengths and concerns separately. Flammer (2013) finds that stock
market reacts to both eco-friendly and eco-harmful events, but that the reaction to eco-friendly events is larger in absolute terms than
the reaction to eco-harmful events. Kruger (2015) finds that there is a negative stock market reaction to both positive and negative
community and environmental CSR announcements, but that the reaction to positive announcements is less negative. This evidence

8
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

suggests possible heterogeneity of the effect of environmental CSR strengths and concerns on product market perception.
Our results suggest that community and environmental CSR strengths significantly positively affect product market perception.
The coefficient estimate for concerns is negative, but not significant at conventional levels. When examining community and en-
vironmental CSR separately, we find that both community and environmental strengths are positively linked to product market
perception, while both community and environmental concerns variables are not significant.
In untabulated tests, we examine the relation between product market perception and other components of CSR: human rights,
employee friendliness, diversity, and corporate governance. The coefficients on employee friendliness and corporate governance are
positive, but not statistically significant. The coefficient on human rights is negative and significant, while coefficient on diversity is
negative but statistically not significant. One possible explanation for this result is that resources spent on human rights and diversity
concerns are viewed as resources which do not impact product market perception.

4.2. Endogeneity concerns

One concern about the findings in the section above is that the effect is driven by endonegeity issues. One possible issue is reverse
causality, since firms with strong product market perception can afford to spend more on CSR activities such as giving back to the
community and protecting the environment. Furthermore, our results might suffer from omitted variables bias. There may be un-
observable characteristics that affect both CSR and product market perception. For example, firms with strong culture and man-
agement might invest more in CSR and in enhancement of product market perception. We address these concerns in two ways. First,
we run an instrumental variables regression. Second, we use a BP Oil Spill as a quasi-natural experiment.
Our first approach is to use instrumental variables. Recent literature suggests that CSR activities around the location of the firm
have a significant effect on the firm's CSR activities through knowledge spillovers and institutional pressures (Husted et al. (2016)).
Thus, we expect the firm's CSR activities to be affected by environmental and community related activities around the firm's location.
In this spirit, we use two instruments in our IV analyses. First, we use per capita CO2 emissions from fossil fuel combustion in the state
where the firm is headquartered. We collect this data from the Environmental Protection Agency's website (epa.gov). We expect firms
located in areas where CO2 emissions are higher to have lower CSR. Second, we use the percentage of state's population that
volunteer for non-profit and community organizations as an instrument. We gather this data from The Corporation for National and
Community Service, which uses Current Population Survey by US Census Bureau to compile this data. We expect volunteerism rates
within a state to be positively related to our CSR measure. However, we do not expect CO2 emissions and volunteer rates in
company's headquartered state to have a direct effect on product perception. In these regressions, we also control for state char-
acteristics such as state's population, median household income, and unemployment rate since our instrumental variables are state-
based.
We present the results of the instrumental variable regressions in Table 4. We find that CO2 emissions is significantly negatively,
and volunteerism rate in the firm's headquartered state is significantly positively related to environmental and community CSR. The
first-stage F-statistics is about 19, suggesting that our instrumental variables are not weak. P-values of overidentification tests
(Sargan's (1958) and Basmann's (1960) chi-squared tests) are not significant suggesting that our instruments are valid. Further,
Wooldridge's (1995) robust score test rejects the null hypothesis that our measure of environmental and community CSR is exogenous
in the model.
Column (2) of Table 4 show that the coefficient estimate for our environmental and community CSR measure remains significant
and positive in the second-stage of the instrumental variable regression, indicating that environmental and community related
activities of firms have a positive impact on product perception.
Our second approach is to use the BP Oil Spill as a quasi-natural experiment. In 2010, Deepwater Horizon oil rig operated by
British Petroleum (BP) exploded, causing one of the largest marine oil spills.13 The event had devastating effects on marine life in the
Gulf of Mexico and is considered one of the largest environmental disasters in U.S. history. The event also attracted attention to
energy firms and their impact on the environment. Our original hypotheses were developed on the basis of competitive differentiation
and signaling value of CSR activities. Thus, we expect firms which have successfully differentiated themselves from competition with
their CSR activities to have better product market perception relative to competitors after an exogenous negative shock to the
industry's CSR perception, such as the BP Oil spill. We also expect these effects to be seen only in energy industry because BP Oil spill
shock primarily affected energy industry.
In Table 5, we examine the effect of this event on product market perception of energy firms. We find that the impact of
environmental and community related CSR activities is stronger after the spill for energy firms.14 This result suggests that these
activities become more important after public relations shock to the firms in the energy industry. In placebo tests, reported in Table 5,
we do not find a significant effect of environmental and community related CSR activities on brand-value after BP oil spill for other
industries.
Overall, the results from the instrumental variable regressions and a quasi-natural experiment analysis indicate that the positive
effect of CSR on product market perception is robust to endogeneity concerns.

13
Bialkowski and Starks (2016) also uses BP Oil Spill as an exogenous event to study the effect of corporate environmental failures on Socially
Responsible Investment (SRI) fund flows. Their results show higher inflows to SRI funds compared to conventional funds following the BP Oil Spill.
14
We use the lagged value of community and environment related CSR measure (before oil spill) in our analysis to account for the possibility that
firms can increase these activities after the oil spill.

9
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 4
Product market perception and community and environmental CSR – IV regressions.
(1) (2)

First-stage Second-stage
Community/ Product Perceptiont+1
Environmental CSR

Community/Environmental CSR 3.163**

(0.042)
Log (Assets) 0.043*** 0.038
(0.000) (0.571)
MB ratio 0.001 0.089***
(0.885) (0.000)
Leverage −0.011 −0.074
(0.497) (0.302)
ROA 0.198*** −0.791*
(0.006) (0.056)
Advertising/Sales 0.527*** −0.780
(0.005) (0.562)
R&D/Sales −0.004 −0.208
(0.977) (0.771)
SG&A/Sales 0.116* 0.002
(0.057) (0.995)
Managerial Ability −0.064*** 0.237*
(0.001) (0.052)
Litigation 0.004 −0.033
(0.799) (0.594)
Restatement 0.023* −0.042
(0.082) (0.531)
Dividend Cut 0.011 −0.077
(0.359) (0.155)
Organizational Capital −0.009 0.011
(0.552) (0.894)
State Population 0.006 0.010
(0.666) (0.834)
State Unemployment −0.035 0.157
(0.402) (0.323)
State Median Household Income −0.048 −0.526*
(0.519) (0.077)
State Volunteer Rate 0.171***
(0.004)
State CO2 Emissions −0.058**
(0.028)
First-stage F-stat 18.77
p-value F-stat 0.000***
Sargan score Chi2 0.816
p-value Sargan score 0.366
Basmann Chi2 0.793
p-value Basmann 0.373
Observations 2476 2476
Year dummies Yes Yes
Industry dummies Yes Yes

This table reports the instrumental variable regression results of Community and Environmental CSR measures on Product
Market Perception. Product Perception is an average of standardized values of the following: Relevance, Knowledge,
Distinctive, Unique, Dynamic, Innovative, Leader, Reliable, High quality, Trustworthy. Community/Environmental CSR is the
difference between community and environmental CSR strengths and concerns from MSCI ESG Research database. Details of
the calculation of these measures are described in the text. Log (Assets) is the logarithm of total assets. MB ratio is the ratio of
market value of assets to book value of assets. Leverage is the ratio of the sum of current liability and long-term debt to total
assets. ROA is the ratio of operating income to total assets. Advertising/Sales is the ratio of advertising expense to sales. R&D/
Sales is the ratio of research and development expense to sales. SG&A/Sales is the ratio of selling, general, and administrative
expenses to sales. Managerial Ability is the measure by Demerjian et al. (2012). Litigation, Restatement, and Dividend Cut are
dummy variables that equal one if the company faced class action lawsuits, restated its financials, or cut its dividend for that
year, respectively. Organizational Capital is calculated following Eisfeldt and Papanikolaou (2013) and scaled by total assets.
State Volunteer Rate is the natural logarithm of percentage of headquartered state’s population that volunteer for non-profit
and community organizations. State CO2 Emissions is the natural logarithm of per capita CO2 emissions from fossil fuel
combustion in the state that the firm is headquartered. State Population, State Unemployment, State Median Household Income
are the natural logarithms of headquartered state’s population, unemployment rate, and median household income level,
respectively. All models include Fama-French (1997) acquirer industry dummies, year dummies, and a constant. p-values
based on robust standard errors clustered at the firm level are in parentheses. *, **, and *** denote statistical significance at
10%, 5%, and 1% levels, respectively.

10
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

4.3. CSR and product market perception in competitive industries, and in differentiated versus standardized goods industries

In Table 6, we examine whether CSR has a greater effect on product market perception in competitive industries, and in dif-
ferentiated versus standardized goods industries.15 We find that the interaction between CSR and Competitive industry dummy is
positive and significant in the regression on Product Perception (Table 6 Column 1).16 This result provides support for the Fisman et al.
(2006) argument that CSR serves as a source of product differentiation in competitive industries and lends support to hypothesis H1A.
We also find support for the prediction that CSR is associated with more positive product market perception in standardized rather
than differentiated product industries. Standardized goods have lower switching costs and substitution elasticities. Therefore, the
impact of CSR on product market perception is likely stronger in standardized products industries. As shown in Column 2 of Table 6,
while both the CSR variable and the differentiated goods industry variable have positive and significant coefficients, their interaction
is negative and significant. This result supports hypothesis H1B.
An alternative way to distinguish differentiated goods from standardized goods is by the amount of research and development (R&
D) spent by the company. Companies with standardized goods are likely to have small levels of R&D. We include in our model a
dummy variable that takes the value of one if the firm has not reported R&D expenditure for that year, and zero otherwise, and
interact it with the CSR variable (Table 6, Column 3). We find that the variable No R&D is negative and statistically significant,
indicating a negative association with product market perception. However, the interaction between the No R&D dummy and CSR is
positive and significant, suggesting that CSR has a greater positive effect on product market perception for standardized goods.

4.4. CSR and product quality

We have argued that CSR serves as a product differentiation strategy and improves perceived product quality. Our data allows us
to create a more precise measure of product quality, which includes only a subsection of attributes in our product market perception
measure. We construct a Quality measure, which includes the attributes of the BAV database Leader, Reliable, High Quality, and
Trustworthy. The measure Other includes the attributes Relevance, Knowledge, Distinctive, Unique, Dynamic, and Innovative.
Table 7 replicates Table 6 separately for Quality and Other. We find that the coefficient on CSR is positive and significant for both
measures and that the coefficient estimates of CSR for Quality measures are significantly higher than that of Other measures, in-
dicating that CSR has a greater impact on quality-related aspects of product market perception (Table 7, Columns (1) and (2)). We
find that the coefficient estimates for the interaction between CSR and the Competitive Industry dummy is significantly higher for
Quality than Other (Table 7, Columns (3) and (4)). Further, the interaction between CSR and the Differentiated goods industry dummy
and No R&D dummy is significant only for Quality, and the coefficient of this estimate is significantly higher than in the models with
Other as the dependent variable. The coefficient on CSR remains significant in all models.
Overall, we find that our results of the effect of CSR on product market perception are stronger for product quality than for other
attributes of product market perception. Furthermore, the results are stronger in competitive industries and for standardized goods.
These results provide further support to our arguments that CSR may signal product quality to the customers, and can be used as a
tool for product differentiation.

4.5. CSR, product market perception, and firm value

Our results, so far, suggest that firms that invest in certain CSR activities benefit in terms of improved product market perception.
Thus, one channel through which CSR can improve firm performance is through improving its product market perception. We
examine whether CSR affects firm value directly and indirectly through product market perception. Several papers suggest an indirect
link between CSR and firm value. Luo and Bhattacharya (2006) propose a mediating effect of customer satisfaction while Servaes and
Tamayo (2013) suggest that customer awareness mediates the link between CSR and firm value. Our paper is the first to examine the
role of product market perception as a channel through which CSR affects firm value.
We focus on two measures of firm value. The first measure is Tobin's Q, which has been widely used in the literature (e.g., Servaes
and Tamayo, 2013). Our results show that product market perception is significantly positively related to contemporaneous and one-
year ahead Tobin's Q (Table 8, Columns 2 and 4).17 One standard deviation change in product market perception increases one-year
ahead Tobin's Q by about 5.8%. Consistent with prior literature we find that coefficient on CSR is positive and significant when
product market perception is excluded (Table 8, Columns (1) and (3)), but becomes insignificant when product market perception is
included (Table 8, Columns (2) and (4)). These results suggests that CSR affects firm value only indirectly through its effect on
product market perception, as predicted in hypothesis H2. We discuss mediation analysis in the next section. Our results are similar to

15
Following Giannetti et al. (2011), differentiated product industries are: furniture and fixture; printing and publishing; rubber and plastic
products; stone, glass, and clay products; fabricated metal products; machinery; electrical equipment; transportation equipment; instruments;
miscellaneous products.
16
We define competitive industries using historical fitted SIC-based HHI data http://hobergphillips.usc.edu/industryconcen.htm. This data is
available only through 2005. For the rest of the years, we calculate HHI based on 3-digit industry and define firms in the bottom 25th percentile as
competitive. The results of the tests with HHI based on 3-digit SIC are similar but weaker.
17
We have slightly higher number of observations in these tests, because, consistent with previous literature that use mediation analysis, we use
contemporaneous values of both Community/Environment CSR and Product Perception. In our earlier tests, we use contemporaneous value of the CSR
measure, and one-year ahead value of our product market perception measure.

11
 K.S. Bardos, et al.

Table 5
Product market perception and community and environmental CSR – evidence from BP oil spill.
Industry dummy = 1 Industry dummy = 1 Industry Industry dummy = 1 Industry dummy = 1 Industry dummy = 1 Industry Industry dummy = 1
for Energy for Consumer Goods dummy = 1 for for Telecom for Bus. Eq. for Manuf,-Chem, dummy = 1 for for Other
Shops Health

Product perceptiont+1 Product perception t+1 Product perception Product perception Product perception Product perception Product perception Product perception t+1

t+1 t+1 t+1 t+1 t+1

Industry x Post-2010 x 1.059** −0.126 −0.086 −1.209 −0.696 0.080 −1.276 −1.481
Com./Env. CSRt-1 (0.029) (0.838) (0.905) (0.516) (0.246) (0.863) (0.191) (0.360)
Industry x Post-2010 0.132 −0.065 0.102* 0.310* 0.225** −0.203*** 0.158 −0.104
(0.405) (0.336) (0.096) (0.065) (0.018) (0.005) (0.261) (0.284)
Industry x Com./Env. −1.035** −0.061 0.066 0.826 1.113* −0.353 1.307 1.430

12
CSRt-1 (0.041) (0.921) (0.933) (0.666) (0.093) (0.511) (0.224) (0.421)
Post-2010 x Com./Env. −0.704*** −0.585** −0.600*** −0.584*** −0.654*** −0.717*** −0.539** −0.565**
CSRt-1 (0.004) (0.014) (0.009) (0.008) (0.005) (0.007) (0.021) (0.011)
Com./Env. CSRt-1 0.903*** 0.820*** 0.827*** 0.811*** 0.723*** 1.013*** 0.735*** 0.765***
(0.001) (0.003) (0.002) (0.001) (0.007) (0.000) (0.005) (0.002)
Controls Yes Yes Yes Yes Yes Yes Yes Yes
Year dummies Yes Yes Yes Yes Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes Yes Yes Yes Yes

This table shows the effect of BP oil spill on product market perception for energy firms and for firms in other industries. Industry is the Fama-French 12 industry reported in the first row. Post-2010 is a
dummy variable that equals one if the year is after 2010, when the BP oil spill happened. Product Perception is an average of standardized values of the following: Relevance, Knowledge, Distinctive,
Unique, Dynamic, Innovative, Leader, Reliable, High quality, Trustworthy. Community/Environmental CSR is the difference between community and environmental CSR strengths and concerns from MSCI
ESG Research database. Details of the calculation of these measures are described in the text. All models include control variables (as in Table 3 Column 1), year, and industry dummies, and a constant. p-
values based on robust standard errors clustered at the firm level are in parentheses. *, **, and *** denote statistical significance at 10%, 5%, and 1% levels, respectively.
Journal of Corporate Finance 62 (2020) 101588
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 6
Product market perception and community and environmental CSR – competitive industries, differentiated vs. standardized goods.
(1) (2) (3)

Product Perceptiont+1 Product Perceptiont+1 Product Perceptiont+1

Community/Environmental CSR*Competitive industry 0.788*

(0.056)
Competitive industry −0.255***
(0.000)
Community/Environmental CSR* Differentiated goods industry −0.390*
(0.071)
Differentiated goods industry 0.439***
(0.000)
Community/Environmental CSR #No R&D 0.388*
(0.063)
No R&D −0.340***
(0.000)
Community/Environmental CSR 0.359*** 0.616*** 0.264**
(0.001) (0.000) (0.019)
Controls Yes Yes Yes
Year dummies Yes Yes Yes
Industry dummies Yes No Yes
Observations 2505 2505 2505
R-squared 0.471 0.296 0.480

This table reports the OLS regression results of Product Market Perception on interactions of Community and Environmental CSR with competitive
industry, differentiated goods industry dummy variable and a dummy variable for firms with no R&D expenditure. Product Perception is an average
of standardized values of the following: Relevance, Knowledge, Distinctive, Unique, Dynamic, Innovative, Leader, Reliable, High quality,
Trustworthy. Community/Environmental CSR is the difference between community and environmental CSR strengths and concerns from MSCI ESG
Research database. Details of the calculation of these measures are described in the text. Differentiated goods industries are defined as in Giannetti
et al. (2011): furniture and fixture; printing and publishing; rubber and plastic products; stone, glass, and clay products; fabricated metal products;
machinery; electrical equipment; transportation equipment; instruments; miscellaneous products. No R&D is a dummy variable that takes the value
of one if the firm has not reported R&D expenditure for that year, and zero otherwise. All models include control variables (as in Table 3 Column 1),
year dummies, and a constant. First and third models also include Fama-French (1997) industry dummies. p-values based on robust standard errors
clustered at the firm level are in parentheses. *, **, and *** denote statistical significance at 10%, 5%, and 1% levels, respectively.

those in Servaes and Tamayo (2013), who find that CSR affects firm value only through its interaction with advertising intensity.
However, as discussed earlier, we argue that our product market perception measure is a more direct measure of consumer channel
than advertising intensity. Moreover, we control for advertising expenditures in all our models.
Lys et al. (2015) find that firms undertake CSR expenditures in the current period when they anticipate stronger future financial
performance. To account for the possibility of the endogeneity of CSR and firm performance, we instrument CSR with State Volunteer
Rate and State CO2 Emissions, as we did in our instrumental variable regressions in Table 4. We use the predicted value of Com-
munity/Environment CSR in our Tobin's Q regressions. In untabulated results, the product market perception continues to be positively
related to Tobin's Q.
It is also possible that highly valued firms can afford to spend more on advertising, thus have better product market perception. To
address this endogeneity concern, we also run instrumental variable regressions. In these regressions, we use two variables from BAV
survey that are related to consumer perception of the brand, but are not related to firm's financial policy as instruments. These
variables are percentage of households that responded positively to the questions —the one I prefer to buy/use and —the one I would
never consider to buy/use.18 In unreported results, we find that the instrumental variable one I prefer to buy/use is significantly
positively related, and the one I would never consider to buy/use is negatively related to our product market perception. First-stage F-
statistics is 62, indicating that these instruments are not weak. P-values of overidentification tests (Sargan's (1958) and Basmann's
(1960) chi-squared tests) are not significant suggesting that our instruments are valid. After controlling for endogeneity, product
market perception continues to be positively related to Tobin's Q.19
Finally, we replicate our results using a different measure of firm value: profit margin. It is likely, that improved product market
perception will allow companies to charge higher prices resulting in improved profit margins. As shown in Table 8, Columns (5)
through (8), the coefficient on Product Perception is still positive and statistically significant. The coefficient on CSR is insignificant in
any of these models.

18
Larkin (2013) also uses similar instruments in an earlier unpublished version of her paper.
19
We also run Tobin's Q regressions with predicted values of both product market perception and community and environment related CSR
measure. In unreported tests we continue to find that product market perception is positively related to Tobin's Q.

13
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 7
Quality and community and environmental CSR – competitive industries, differentiated vs. standardized goods.
(1) (2) (3) (4) (5) (6) (7) (8)

Qualityt+1 Othert+1 Qualityt+1 Othert+1 Qualityt+1 Othert+1 Qualityt+1 Othert+1

Community/Environmental CSR 0.429*** 0.367*** 0.371*** 0.351*** 0.710*** 0.554*** 0.247* 0.275**
(0.002) (0.001) (0.006) (0.001) (0.000) (0.000) (0.083) (0.011)
Community/Environmental CSR*Competitive 1.411** 0.373
industry (0.011) (0.324)
Competitive industry −0.364*** −0.183***
(0.000) (0.004)
Community/Environmental CSR* Differentiated −0.634** −0.227
goods industry (0.028) (0.289)
Differentiated goods industry 0.548*** 0.367***
(0.000) (0.000)
Community/Environmental CSR *No R&D 0.616*** 0.236
(0.009) (0.277)
No R&D −0.337*** −0.342***
(0.001) (0.000)
Controls Yes Yes Yes Yes Yes Yes Yes Yes
Year dummies Yes Yes Yes Yes Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes No No Yes Yes
Observations 2505 2505 2505 2505 2505 2505 2505 2505
R-squared 0.552 0.380 0.564 0.386 0.353 0.246 0.564 0.401
Coefficient test p-value 0.09* 0.005*** 0.044** 0.011***

This table reports the OLS regression results of two components of Product Market Perception (Quality and Other) on interactions of Community and
Environmental CSR with competitive industry, differentiated goods industry dummy variable and a dummy variable for firms with no R&D ex-
penditure. Quality includes Leader, Reliable, High quality, and Trustworthy. Other includes Relevance, Knowledge, Distinctive, Unique, Dynamic,
Innovative. Community/Environmental CSR is the difference between community and environmental CSR strengths and concerns from MSCI ESG
Research database. Details of the calculation of these measures are described in the text. Differentiated goods industries are defined as in Giannetti
et al. (2011): furniture and fixture; printing and publishing; rubber and plastic products; stone, glass, and clay products; fabricated metal products;
machinery; electrical equipment; transportation equipment; instruments; miscellaneous products. No R&D is a dummy variable that takes the value
of one if the firm has not reported R&D expenditure for that year, and zero otherwise. All models include control variables (as in Table 3 Column 1),
year dummies, and a constant. All models, except fifth and sixth, also include Fama-French (1997) industry dummies. p-values based on robust
standard errors clustered at the firm level are in parentheses. The last row reports p-value for the test that the coefficient estimate for CSR variable
and its interaction is larger for quality measure than other measures. *, **, and *** denote statistical significance at 10%, 5%, and 1% levels,
respectively.
Coefficient estimates of interest are in bold.

4.6. Mediation analysis

To more formally test whether product market perception is a channel through which CSR affects firm value we perform med-
iation analysis.20 Traditional mediation analysis was developed by Preacher and Hayes (2004) (known as causal steps approach
popularized by Baron and Kenny (1986)). This analysis would imply the following relations for our research questions:

Firm Performance = i1 + c CSR (2)

Product Market Perception = i2 + a CSR (3)

Firm Performance = i3 + c CSR + b Product Market Perception (4)

In this set-up, product market perception is the mediator. When the effect of CSR on Firm Performance is decreased, partial
mediation is said to have occurred.21 In our analysis X is CSR and Y is the firm performance measures.
Column (1) of Table 9 estimates eq. (1) and finds a positive and significant coefficient on CSR. Table 3 estimates eq. (1) and
establishes a positive coefficient on CSR, which is robust to endogeneity concerns. Column (2) of Table 9 estimates eq. (4) and finds
that the coefficient on product market perception is positive and significant, while the coefficient on CSR becomes insignificant.
Coefficient on CSR is reduced by 43% due to the addition of product market perception, suggesting that partial mediation has
occurred. These results show that product market perception is a channel through which CSR affects firm value.

20
This methodology has been widely used in many disciplines. For finance applications see Ferris et al. (2017) and Fedaseyeu et al. (2018)
21
Baron and Kenny (1986) require for the coefficient c to be different from zero. However, more recent literature relaxes this assumption (Collins
et al. (1998), MacKinnon (2000), and Shrout and Bolger (2002). Shrout and Bolger (2002) say that they “support recommendations to set aside the
first step of Baron and Kenny's (1986) classic approach… we recommend that the mediation analysis proceed on the basis of the strength of the
theoretical arguments rather than on the basis of the statistical test of X on Y.”

14
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 8
Product market perception, CSR, and firm value.
(1) (2) (3) (4) (5) (6) (7) (8)

Qt Qt Qt+1 Qt+1 Profit margint Profit margint Profit Profit margint+1

margint+1

Community/Environmental 0.086* 0.042 0.086* 0.049 0.010 −0.001 0.014 0.006

CSRt (0.070) (0.340) (0.087) (0.301) (0.489) (0.957) (0.305) (0.670)
Product Perceptiont 0.105*** 0.090*** 0.026*** 0.018**
(0.000) (0.000) (0.002) (0.014)
Log (Assets) −0.034*** −0.048*** −0.026*** −0.038*** 0.009*** 0.006* 0.012*** 0.010***
(0.000) (0.000) (0.002) (0.000) (0.002) (0.094) (0.000) (0.001)
Leverage 0.122*** 0.129*** 0.159*** 0.166*** −0.031*** −0.030*** −0.006 −0.005
(0.005) (0.001) (0.000) (0.000) (0.001) (0.002) (0.600) (0.695)
ROA 1.456*** 1.422*** 1.266*** 1.237*** 0.487*** 0.479*** 0.390*** 0.384***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Advertising/sales 0.085 −0.023 0.239 0.150 0.061 0.035 −0.033 −0.051
(0.838) (0.951) (0.530) (0.668) (0.540) (0.712) (0.725) (0.573)
R&D/sales 1.068*** 1.006*** 1.033*** 0.979*** −0.145 −0.160 −0.069 −0.080
(0.003) (0.003) (0.002) (0.002) (0.376) (0.302) (0.570) (0.486)
SG&A/sales 0.323** 0.297** 0.310** 0.287** −0.012 −0.018 0.054 0.049
(0.026) (0.030) (0.017) (0.020) (0.759) (0.632) (0.143) (0.171)
Managerial ability 0.064** 0.050** 0.051* 0.039 −0.004 −0.007 −0.020* −0.023**
(0.014) (0.047) (0.054) (0.136) (0.736) (0.519) (0.056) (0.037)
Litigation 0.003 0.003 0.030 0.029 −0.048* −0.048* 0.002 0.002
(0.920) (0.922) (0.323) (0.325) (0.077) (0.076) (0.879) (0.891)
Restatement −0.024 −0.026 −0.036 −0.038 −0.001 −0.001 −0.009 −0.010
(0.296) (0.226) (0.135) (0.104) (0.940) (0.890) (0.330) (0.305)
Dividend Cut −0.058*** −0.050*** −0.046*** −0.040** −0.022** −0.020** −0.024** −0.023**
(0.001) (0.003) (0.008) (0.018) (0.023) (0.030) (0.016) (0.020)
Organizational Capital 0.090** 0.088** 0.072* 0.071* −0.013 −0.013 −0.007 −0.007
(0.024) (0.017) (0.099) (0.092) (0.251) (0.231) (0.623) (0.608)
Year dummies Yes Yes Yes Yes Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes Yes Yes Yes Yes
Observations 2696 2696 2647 2647 2696 2696 2646 2646

This table reports OLS regressions. The dependent variables are Tobin's Q, which is the log of one plus the ratio of market value of assets to book
value of assets, and Profit Margin, which is net income divided by sales. Community/Environmental CSR is the difference between community and
environmental CSR strengths and concerns from MSCI ESG Research database. Details of the calculation of these measures are described in the text.
Log (Assets) is the logarithm of total assets. Leverage is the ratio of the sum of current liability and long-term debt to total assets. ROA is the ratio of
operating income to total assets. Advertising/Sales is the ratio of advertising expense to sales. R&D/Sales is the ratio of research and development
expense to sales. SG&A/Sales is the ratio of selling, general, and administrative expenses to sales. Managerial Ability is the measure by Demerjian
et al. (2012). Litigation, Restatement, and Dividend Cut are dummy variables that equal one if the company faced class action lawsuits, restated its
financials, or cut its dividend for that year, respectively. Organizational Capital is calculated following Eisfeldt and Papanikolaou (2013) and scaled
by total assets. All models include Fama-French (1997) industry dummies, year dummies, and a constant. p-values based on robust standard errors
clustered at the firm level are in parentheses. *, **, and *** denote statistical significance at 10%, 5%, and 1% levels, respectively.

5. Conclusion

Using a large and proprietary database of product market perception, we are able to directly examine how CSR activities of the
firm impact product market perception. We find that community and environmental CSR improve product market perception. This
result suggests that the customer channel is an important channel through which CSR affects the firm. Using a quasi-natural ex-
periment and instrumental variable regressions, we show that the positive relation between environmental and community CSR and
product market perception is robust to controlling for endogeneity.
We also find that the effect of CSR on product market perception is stronger for CSR strengths rather than concerns. The effect of
CSR on product market perception is positive and highly significant for both community and environmental strengths. The effect is
negative, but not significant for community and environmental concerns. This result suggests that possibly there is more customer
awareness about CSR strengths than concerns. We find that other CSR components (employee friendliness, diversity, and corporate
governance) are not associated with product market perception. In addition, we find that the association between CSR and product
market perception is weaker for differentiated goods industries compared to standardized goods industries, and stronger for com-
petitive industries. Furthermore, this association is more pronounced for product quality attributes than for other attributes of the
product market perception.
We also contribute to the literature examining the relation between CSR and firm value. We find that product market perception is
significantly positively related to firm value, but CSR is not when product market perception is included in the regression. This result
is consistent with the stakeholder view of CSR and suggests that CSR increases firm value indirectly by improving product market
perception.

15
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Table 9
Mediation analysis.
(1) (2)

Qt+1 Qt+1

Product Perceptiont 0.090***

(0.000)
Community/Environment CSRt 0.086* 0.049
(0.087) (0.301)
Controls Yes Yes
Year dummies Yes Yes
Industry dummies Yes Yes
Coefficient test p-value 0.007***

This table show the results of mediation analysis. The first column shows the effect of Community/
Environment CSR on Tobin's Q. The second column shows the effect when Product Perception is included
in the specification. These equations have been jointly estimated. The last row shows the p-value for
the test that the coefficient estimates for Community/Environment CSR in the two specifications are
significantly different from each other. All model include control variables (as in Table 8), year, Fama-
French (1997) industry dummies, and a constant. p-values based on robust standard errors clustered at
the firm level are in parentheses. *, **, and *** denote statistical significance at 10%, 5%, and 1%
levels, respectively.

Overall, we find that visible CSR, such as community and environmental, positively affects firm value by improving product
market perception. The results suggest that the customer is an important stakeholder through which CSR creates firm value.

Acknowledgement

We would like to thank the editor, William Megginson, two anonymous reviewers, Kiyoung Chang, Karthik Krishnan, Nathan
Mislang, Rebekah Moore, Chi Wan, and the participants at the 2019 Corporate Social Responsibility across the Atlantic Conference,
2019 European Financial Management Association Annual Meeting, 2017 Financial Management Association Annual Meeting, and
2017 Southern Finance Association Annual Meeting for their helpful suggestions. We thank BAV Consulting for providing us the
brand data. The paper was previously circulated with the title “The Mediating Role of Brand Asset on the Valuation of Corporate
Social Responsibility”.

Appendix A. Appendix

We follow Larkin (2013) and identify the brand that is closest to the corporation's name as follows:Monobrands: For monobrands,
brand represents all or most of the firm's business (e.g., Starbucks,
FedEx, Delta Airlines). Since the company and brand names are the same, we use the brand perception scores of the brand for the
company.
Corporate brands: The corporate name is dominant in the brand name (e.g., Apple, Colgate). In this case, we use brand perception
scores for the company name. For example, we use the brand scores of Apple, instead of iPhone or iPad.
House of brands: For house of brands strategy, the corporation does not use its corporate name in brands. For example, Procter and
Gamble owns Olay, Tide, Crest toothpastes, etc. Although in these cases it might be more accurate to use the weighted average of the
brand perception scores of the company, implementing this approach is challenging. Ideally, we should use a weighted average
(based on revenues or profits) brand scores of each product the company owns. However, companies do not typically report the
revenues for each product. Furthermore, BAV data does not include all the brands that a company owns. Fortunately, BAV not only
surveys the brands but the company as well. For example, BAV data includes brand perception scores for Procter and Gamble as well
as some of its brands. In our study, we use the brand scores of the company (in this example Procter and Gamble) rather than the
average score of the brands of the company.
Mixed brands: Sometimes corporations use their name for some of their products, but not for others. For example, Gap Inc. is the
owner of Gap Stores, Banana Republic, Old Navy, and Athleta brands. In these case, we use the brand perception scores of the brand
that is most similar to the name of the company (e.g., Gap for Gap Inc.).

References

Adhikari, B.K., 2016. Causal effect of analyst following on corporate social responsibility. J. Corp. Finan. 41, 201–216.
Albuquerque, R., Durnev, A., Koskinen, Y., 2017. Corporate Social Responsibility and Firm Risk: Theory and Empirical Evidence. European Corporate Governance
Institute (ECGI) - Finance Working Paper No. 359/2013. Available at SSRN. https://ssrn.com/abstract=1961971.
Anderson, E.W., Sullivan, M.W., 1993. The antecedents and consequences of customer satisfaction for firms. Mark. Sci. 12 (2), 125–143.
Bagnoli, M., Watts, S.G., 2003. Selling to socially responsible consumers: competition and the private provision of public goods. J. Econ. Manag. Strateg. 12 (3),
419–445.
Barnett, M.L., 2007. Stakeholder influence capacity and the variability of financial returns to corporate social responsibility. Acad. Manag. Rev. 32 (3), 794–816.

16
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Barney, J., 1991. Firm resources and sustained competitive advantage. J. Manag. 17 (1), 99–120.
Baron, D., 2001. Private politics, corporate social responsibility and integrated strategy. J. Econ. Manag. Strateg. 10, 7–45.
Baron, R.M., Kenny, D.A., 1986. Themoderator-mediator variable distinction in social psychological research: conceptual, strategic and statistical considerations. J.
Pers. Soc. Psychol. 51, 1173–1182.
Basmann, R.L., 1960. On finite sample distributions of generalized classical linear identifiability test statistics. J. Am. Stat. Assoc. 55, 650–659.
Berland, Penn Schoen, 2010. Corporate Social Responsibility Branding Survey. Penn Schoen Berland, New York.
Bhandari, A., Javakhadze, D., 2017. Corporate social responsibility and capital allocation efficiency. J. Corp. Finan. 43, 354–377.
Bhattacharya, C.B., Sen, S., 2004. Doing better at doing good: when, why, and how consumers respond to corporate social initiatives. Calif. Manag. Rev. 47, 9–24.
Bialkowski, J., Starks, L.T., 2016. SRI Funds: Investor Demand, Exogenous Shocks and ESG Profiles, Working Papers in Economics 16/11. University of Canterbury,
Department of Economics and Finance.
Boone, A., Uysal, V.B., 4 Sep 2018. Reputational concerns in the market for corporate control. J. Corp. Finan (In press).
Castaldo, S., Perrini, F., Misani, N., Tencati, A., 2009. The missing link between corporate social responsibility and consumer trust: the case of fair trade products. J.
Bus. Ethics 84 (1), 1–15.
Collins, L.M., Graham, J.W., Flaherty, B.P., 1998. An alternative framework for defining mediation. Multivar. Behav. Res. 33, 295–312.
Cronqvist, H., Heyman, F., Nilsson, M., Svaleryd, H., Vlachos, J., 2009. Do entrenched managers pay their workers more? J. Financ. 64, 309–339.
Demerjian, P., Lev, B., McVay, S., 2012. Quantifying managerial ability: a new measure and validity tests. Manag. Sci. 58 (7), 1229–1248.
Deng, X., Kang, J.-K., Low, B.S., 2013. Corporate social responsibility and stakeholder value maximization: evidence from mergers. J. Financ. Econ. 110, 87–109.
Eisfeldt, A.L., Papanikolaou, D., 2013. Organization capital and the cross-section of expected returns. J. Financ. 68 (4), 1365–1406.
Erhemjamts, O., Li, Q., Venkateswaran, A., 2013. Corporate social responsibility and its impact on firms' investment policy, organizational structure, and performance.
J. Bus. Ethics 118 (2), 395–412.
Fatemi, A., Fooladi, I., Tehranian, H., 2015. Valuation effects of corporate social responsibility. J. Bank. Financ. 59, 182–192.
Fedaseyeu, V., Linck, J.S., Wagner, H.F., 2018. Do qualifications matter? New evidence on board functions and director compensation. J. Corp. Finan. 48, 816–839.
Ferrell, A., Liang, H., Renneboog, L., 2016. Socially responsible firms. J. Financ. Econ. 122 (3), 585–606.
Ferris, S.P., Javakhadze, D., Rajkovic, T., 2017. CEO social capital, risk-taking and corporate policies. J. Corp. Finan. 47, 46–71.
Fisman, R., Heal, G., Nair, V.B., 2006. A Model of Corporate Philanthropy. Working Paper. Columbia University and University of Pennsylvania.
Flammer, C., 2013. Corporate social responsibility and shareholder reaction: the environmental awareness of investors. Acad. Manag. J. 56 (3), 758–781.
Flammer, C., Hong, B., Minor, D., 2017. Corporate Governance and the Rise of Integrating Corporate Social Responsibility Criteria in Executive Compensation:
Effectiveness and Implications for Firm Outcomes. Working paper. Boston University, University of Western Ontario and Northwestern University.
Fombrun, C., Shanley, M., 1990. What's in a name? Reputation building and corporate strategy. Acad. Manag. J. 33 (2), 233–258.
Freeman, R.E., 1984. Strategic Management: A Stakeholder Perspective. Pitman Publishing Inc., Boston, MA.
Freeman, R.E., McVea, J., 2001. A stakeholder approach to strategic management. In: Hitt, M.A., Freeman, R.E., Harrison, J.S. (Eds.), Handbook of Strategic
Management. Blackwell, Oxford, pp. 189–207.
Freeman, R.E., Wicks, A.C., Parmar, B., 2004. Stakeholder theory and the corporate objective revisited. Organ. Sci. 15, 364–369.
Frieder, L., Subrahmanyam, A., 2005. Brand perception and the market for common stock. J. Financ. Quant. Anal. 40, 57–85.
Friedman, M., 1970. The social responsibility of business is to increase its profits. New York Times Magazine 13, 122–126 September.
Friedman, M., 1998. The social responsibility of business is to increase its profits. In: Pincus, L.B. (Ed.), Perspectives in Business Ethics. McGraw- Hill, Singapore, pp.
246–251.
Galbreath, J., Shum, P., 2012. Do customer satisfaction and reputation mediate the CSR FP link? Evidence from Australia. Aust. J. Manag. 37 (2), 211–229.
Garcia-Castro, R., Ariño, M.A., Canela, M.A., 2010. Does social performance really lead to financial performance? Accounting for endogeneity. J. Bus. Ethics 92 (1),
107–126.
Giannetti, M., Burkart, M., Ellingsen, T., 2011. What you sell is what you lend? Explaining trade credit contracts. Rev. Finan. Stud. 24, 1261–1298.
Giuli, A.D., Kostovetsky, L., 2014. Are red or blue companies more likely to go green? Politics and corporate social responsibility. J. Financ. Econ. 111, 158–180.
Gregory, A., Whittaker, J., 2013. Exploring the valuation of corporate social responsibility—a comparison of research methods. J. Bus. Ethics 116 (1), 1–20.
Harjoto, M.A., Jo, H., 2011. Corporate governance and CSR nexus. J. Bus. Ethics 100, 45–67.
Helm, S., 2007. One reputation or many? Comparing stakeholders' perceptions of corporate reputation. Corp. Commun. Int. J. 12 (3), 238–254.
Hsu, K.-T., 2012. The advertising effects of corporate social responsibility on corporate reputation and brand equity: evidence from the life insurance industry in
Taiwan. J. Bus. Ethics 109 (2), 189–201.
Hur, W.-M., Kim, H., Woo, J., 2014. How CSR leads to corporate brand equity: mediating mechanisms of corporate brand credibility and reputation. J. Bus. Ethics 125
(1), 75–86.
Husted, B.W., Jamali, D., Saffar, W., 2016. Near and dear? The role of location in CSR engagement. Strateg. Manag. J. 37 (10), 2050–2070.
Jensen, M.C., 2001. Value maximization, stakeholder theory and the corporate objective function. J. Appl. Corp. Finan. 14, 8–21.
Jones, R., 2005. Finding sources of brand value: developing a stakeholder model of brand equity. Brand Manag. 13 (1), 10–32.
Kirmani, A., Rao, A.R., 2000. No pain, no gain: a critical review of the literature on signaling unobservable product quality. J. Mark. 64 (2), 66–79.
Kitzmueller, M., Shimshack, J., 2012. Economic perspectives on corporate social responsibility. J. Econ. Lit. 50 (1), 51–84.
Kotha, S., Rajgopal, S., Rindova, V., 2001. Reputation building and performance: an empirical analysis of the top-50 pure internet firms. Eur. Manag. J. 19 (6),
571–586.
Kruger, P., 2015. Corporate goodness and shareholder wealth. J. Financ. Econ. 115, 304–329.
Lai, C.S., Chiu, C.J., Yang, C.F., Pai, D.C., 2010. The effects of corporate social responsibility on brand performance: the mediating effect of industrial brand equity and
corporate reputation. J. Bus. Ethics 95 (3), 457–469.
Larkin, Y., 2013. Brand perception, cash flow stability, and financial policy. J. Financ. Econ. 110 (1), 232–253.
Lev, B., Radhakrishnan, S., 2005. The Valuation of Organization Capital, in Measuring Capital in the New Economy. University of Chicago Press, Chicago.
Liang, H., Renneboog, L., 2017. On the foundations of corporate social responsibility. J. Financ. 72 (2), 853–910.
Lins, K.V., Servaes, H., Tamayo, A., 2017. Social capital, trust, and firm performance: the value of corporate social responsibility during the financial crisis. J. Financ.
72 (4), 1785–1824.
Luo, X., Bhattacharya, C.B., 2006. Corporate social responsibility, customer satisfaction, and market value. J. Mark. 70 (4), 1–18.
Lys, T., Naughton, J.P., Wang, C., 2015. Signaling through corporate accountability reporting. J. Account. Econ. 60 (1), 56–72.
MacKinnon, D.P., 2000. Contrasts in multiple mediator models. In: Rose, J.S., Chassin, L., Presson, C.C., Sherman, S.J. (Eds.), Multivariate Multivariate Applications in
Substance Use Research. Erlbaum, Mahwah, NJ, pp. 141–160.
Margolis, J.D., Walsh, J.P., 2003. Misery loves companies: rethinking social initiatives by business. Adm. Sci. Q. 48, 268–305.
Margolis, J.D., Elfenbein, H.A., Walsh, J.P., 2007. Does it Pay to Be Good? A Meta-Analysis and Redirection of Research on the Relationship between Corporate Social
and Financial Performance. Working paper. University of Michigan.
McWilliams, A., Siegel, D.S., 2000. Corporate social responsibility and financial performance: correlation or misspecification? Strateg. Manag. J. 21, 603–609.
McWilliams, A., Siegel, D.S., Wright, P.M., 2006. Corporate social responsibility: strategic implications. J. Manag. Stud. 43 (1), 1–18.
Melo, T., Galan, J.I., 2011. Effects of corporate social responsibility on brand value. J. Brand Manag. 18, 423–437.
Mishra, D.R., 2017. Post-innovation CSR performance and firm value. J. Bus. Ethics 140 (2), 285–306.
Mizik, N., Jacobson, R., 2008. The financial value impact of perceptual brand attributes. J. Mark. Res. 45 (1), 15–32.
Orlitzky, M., 2001. Does firm size confound the relationship between corporate social performance and financial performance? J. Bus. Ethics 33, 167–180.
Orlitzky, M., Schmidt, F.L., Rynes, S.L., 2003. Corporate social and financial performance: a meta-analysis. Organ. Stud. 24 (3), 403–441.
Pagano, M., Volpin, P.F., 2005. Managers, workers, and corporate control. J. Financ. 60, 841–868.

17
K.S. Bardos, et al. Journal of Corporate Finance 62 (2020) 101588

Porter, M.E., 1991. America's green strategy. Sci. Am. 264 (4), 96–103.
Porter, M.E., Kramer, M.R., 2006. Strategy and society: the link between competitive advantage and corporate social responsibility. Harv. Bus. Rev. 84 (12), 78–92.
Porter, M.E., Kramer, M.R., 2011. The big idea: creating shared value. Harv. Bus. Rev. 89 (1–2), 62–77.
Preacher, K.J., Hayes, A.F., 2004. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behav. Res. Methods Instrum. Comput. 36 (4),
717–731.
Renneboog, L., Horst, J.T., Zhang, C., 2008. Socially responsible investments: institutional aspects, performance, and investor behavior. J. Bank. Financ. 32 (9),
1723–1742.
Roberts, P.W., Dowling, G.R., 2002. Corporate reputation and sustained superior financial performance. Strateg. Manag. J. 23 (12), 1077–1093.
Saeidi, S.P., Sofian, S., Saeidi, P., Saeidi, S.P., Saaeidi, S.A., 2015. How does corporate social responsibility contribute to firm financial performance? The mediating
role of competitive advantage, reputation, and customer satisfaction. J. Bus. Res. 68 (2), 341–350.
Sargan, J.D., 1958. The estimation of economic relationships using instrumental variables. Econometrica 26, 393–415.
Sen, S., Bhattacharya, C.B., 2001. Does doing good always lead to doing better? Consumer reactions to corporate social responsibility. J. Mark. Res. 38 (2), 225–243.
Servaes, H., Tamayo, A., 2013. The impact of corporate social responsibility on firm value: the role of customer awareness. Manag. Sci. 59 (5), 1045–1061.
Shrout, P.E., Bolger, N., 2002. Mediation in experimental and nonexperimental studies: new procedures and recommendations. Psychol. Methods 7, 422–445.
Siegel, D.S., Vitaliano, D.F., 2007. An empirical analysis of the strategic use of corporate social responsibility. J. Econ. Manag. Strateg. 16 (3), 773–792.
Tian, K.T., Bearden, W.O., Hunter, G.L., 2001. Consumers' need for uniqueness: scale development and validation. J. Consum. Res. 28 (1), 50–66.
Torres, A., Bijmolt, T.H.A., Tribo, J.A., Verhoef, P., 2012. Generating global brand equity through corporate social responsibility to key stakeholders. Int. J. Res. Mark.
29, 13–24.
Van Beurden, P., Gössling, T., 2008. The worth of values — a literature review on the relation between corporate social and financial performance. J. Bus. Ethics 82
(2), 407–424.
Wooldridge, J.M., 1995. Score diagnostics for linear models estimated by two stage least squares. In: Maddala, G.S., Phillips, P.C.B., Srinivasan, T.N. (Eds.), Advances
in Econometrics and Quantitative Economics: Essaysin Honor of Professor C. R. Rao. Blackwell, Oxford, pp. 66–87.

18