Customers Increase Financial Performance of Socially Responsible Firms

Abstract

Previous survey research has documented that consumers place value on socially responsible firms. This support includes the intention to be more loyal to these firms and also the willingness to pay higher prices for their products. Our study connects the customer intentions documented in survey research with actual measures of financial performance from published financial statements. The study uses gross profits scaled by total assets as a proxy for customers’ willingness to pay higher prices and sales increases as a proxy for loyalty. Additionally, the study examines differences in the aforementioned measures between customers in the business-to-business (B2B) and business-to-consumer (B2C) segments. These differences have been documented in studies that suggest customers in these segments value different characteristics of suppliers when making their purchases. Finally, customers must be made aware of a firm’s sustainability practices; therefore, the study looks at three different approaches firms use to communicate the quality of their sustainability practices. These approaches include external assurance of the social responsibility report, the auditor’s review of the firm’s internal controls, and the firm’s advertising intensity. Data used in this study includes financial performance measures of North American firms and corporate social responsibility data from disclosures collected by the Global Reporting Initiative. Using ordinary least squares, the results suggest that customers require some sort of assurance of a company’s socially responsible disclosures when making decisions about whether to support the company.

1. Introduction

A firm’s stakeholders either influence or are influenced by management’s strategies to achieve the firm’s objectives [1,2]. Stakeholder theory focuses on how the firm’s strategies encourage stakeholders in the firm’s value creation [3,4]. This paper focuses on a particular group of stakeholders, customers, and how a firm’s business model creates value for this group of stakeholders [5,6]. The firm’s financial results described its financial value and are relevant for investors as stakeholders. However, non-financial results have been shown to be relevant to stakeholders’ decisions to support the firms’ business models [7]. Numerous studies have confirmed that shareholders and lenders are more willing to support firms that adopt socially responsible and sustainable practices [8,9,10,11]. This support is justified as there is evidence of the value of these practices, as these firms have better returns than the market, better financial efficiency [12], lower cost of capital [13,14], and lower risk [15]. Additionally, Achtenhagen, Melin, & Naldi [16] argue that the financial performance of socially responsible firms is more sustainable. However, the real basis for this support is a willingness on the part of customers to purchase products from sustainable firms. Customer support for firms’ business models is critical, as this support can result in increased revenues, which will improve their financial performance [17]. Without the support from this group of stakeholders, even firms with the best sustainable practices cannot survive. Therefore, customers must see a firm’s sustainable practices as creating sufficient value that will make the firm worthy of their support. The purpose of this study is to extend previous research into stakeholder theory by examining whether customers, as stakeholders, will add value to socially responsible firms by offering their financial support.

Previous studies [18,19,20] have used survey responses to conclude that customers would be more willing to support socially responsible firms. There are two ways in which customers can support socially responsible firms. First, Madden, Roth, & Dillon [21] argue for a “halo” effect, as customers may be more willing to purchase products from firms that disclose their socially responsible actions. The survey research supports this “halo” effect, as customers indicate that they would concentrate their purchases from those firms that are more socially responsible [22,23]. Second, surveys have also indicated customers’ willingness to pay higher prices from socially responsible firms [24]. Each of these types of support can add value to the firm by impacting their financial performance. This research extends these survey-based studies by examining firms’ reported financial performance to determine whether customers actually increase their purchases and pay higher prices from socially responsible firms. The primary contribution of this research is to determine if firms’ corporate social responsibility (CSR) disclosures affect customers’ support and whether/how this adds value to the firm by impacting firms’ financial performance. To investigate this issue, the study examines the relationship between customer-based measures of profitability and CSR reports for North American companies. In addition to examining the impact of CSR reports on customer-related profitability measures, this study also contributes to the research on customer support of socially responsible firms by examining three related issues.

While customers, as a group of stakeholders, may be more loyal and pay premium prices for products from firms that disclose their socially responsible practices, there is a possibility that different customer segments respond differently to these messages. Therefore, as a second contribution, we examine whether there is a differential response to socially responsible firms that deal directly with customers (B2C) versus those that sell to other businesses (B2B) [25,26]. In addition to the differential impact of socially responsible messages on these customer segments, there is also research that suggests that the quality assigned to CSR disclosures is important. Therefore, as a third contribution, we also investigated three issues of CSR report quality.

As a first measure of quality, we examine the Global Reporting Initiatives’ (GRI) suggestion to use application levels to indicate the report’s coverage. The levels are based on the inclusion of important categories of CSR reports. However, as CSR reports filed in accordance with GRI guidelines and the topics included are influenced by the subjectivity of the creator, the report’s quality can be enhanced by some level of assurance [27]. Our second quality measure concerns the type of assurance provided. While assurance provides some indication of the report’s quality, there can be concerns about the firm’s decision to have their CSR report assured. Because CSR reports are voluntary, and firms can purchase assurance separately, there is a possibility for the firm to contract for a level of assurance that is most favorable [28]. Therefore, in addition to direct assurance, we also examine an indirect measure that customers might use to infer the quality of a CSR report.

As an indirect measure of CSR report quality, we examine how a company’s commitment to social responsibility is communicated to customers. Advertising is one such channel and has been shown to legitimize firms’ CSR disclosures as they communicate their intentions to a wider audience [29]. In addition, advertising can also counteract any skepticism customers may have about a firm’s CSR actions [30]. Thus, as a third type of assurance, this study looks at the role of advertising as a communication channel for customers.

Data for this study comes from two sources. Social responsibility reports come from the GRI [31,32]. The financial data comes from Compustat North America. The data is analyzed using ordinary least squares (OLS) and robust regression (RR). Our results show that CSR reports that are assured by an independent assurance provider have a significant positive effect on customer-based profitability. Our findings are important for both regulators and investors, as they indicate that customers use CSR reports in their decisions concerning whether to purchase products from a firm and also in their willingness to pay higher prices. Additionally, the results also show that external assurance and advertising moderate the impact of CSR reports on customer-based profitability positively.

The remainder of the study is organized as follows. Section 2 examines the literature review and presents the hypotheses. Section 3 is about methodology, followed by Section 4, which discusses the results of empirical analyses. Section 5 is about conclusions and implications.

2. Literature Review and Hypotheses Development

2.1. Customers’ Value of Socially Responsible Disclosures

Previous studies document that customers place value on firms’ sustainability actions when considering purchase decisions [11], when supporting investment funds [33], and when finally evaluating firms and their products [21,34]. When firms publish socially responsible and sustainability reports, there is evidence that customers ascribe additional value to the firms’ products [35]. These published reports of firms’ socially responsible and sustainable practices support stakeholder theory, as these disclosures provide stakeholders, including customers, with information about their products, which can satisfy their complex needs and wants [2,36]. Thus, socially responsible and sustainability reports can influence customers’ perceptions of firms’ abilities and incentives to fulfill implied commitments and to provide goods of higher value [37]. Spiller & Belogolova’s [38] survey research documents that when customers view a firm’s products as providing greater value, they are more willing to pay higher prices. In addition, when firms establish a good reputation for social responsibility, this also increases customers’ loyalty [39,40]. This loyalty increases customers’ willingness to purchase products from socially responsible firms [41,42].

There is evidence that customers as a group have intentions to support firms that disclose their CSR and sustainable actions. However, different customer segments may respond differently to CSR and sustainability reports. There are numerous studies that consider customers in either business-to-business (B2B) [25] or business-to-consumer (B2C) segments [26] as responding differently to messages from suppliers. Essentially, the difference between these segments depends on whether the firm’s output is used as an intermediary product or is sold to final customers. Customers in each segment look for different factors, such as the content of their marketing messages, when considering which suppliers to support [43]. Guo, Hwang, & Wang [44] examine the importance of specific advertising to the B2B segment on the firm’s market value and show these customers react to advertising differently. In the e-commerce sector of the B2C segment, customers place a high value on the speed of delivery [45], while in the B2B segment, the overall logistics of the supplier’s delivery system impacts the loyalty of their customers [23]. Du et al. [28] document that in the B2C segment customers also place a high value on website quality and information sharing in their product searches. Bazrafshan et al. [46] also show that information sharing in e-commerce is important in the B2C segment. Han and Lee [47] show that in the B2B segment a firm’s sustainable and socially responsible actions will increase their customers’ trust in their products and messages. Salam, Jahed, & Palmer’s [48] study shows that a firm’s sustainable orientation will increase both customer satisfaction and brand loyalty in the B2B segment. Tzanidis, Magni, Scuotto, & Maalaoui [49] report similar results, and they show that green marketing can increase profits in the B2B segment. Finally, Sharma, Kingshott, Leung, & Malik [50] document the necessity of value co-creation relationships in the B2B segment. These studies indicate the potential benefit and relevance of investigating a possible distinction between these segments in their response to the firm’s CSR reports.

These and other studies on customers’ reactions to firms’ socially responsible and sustainability reports are based on responses to questionnaires or the results of behavioral experiments [17,51,52]. These studies only indicate customers’ intentions or willingness to support socially responsible firms. The gap between intentions and actions has been examined in other settings [53,54]; this study is one of the first to look at customers’ actions and not just their intentions to act. This study explores the link between survey research on customers’ intentions and actual support when firms provide evidence that social responsibility is actually part of their business model.

Therefore, we propose the following hypothesis in null form:

H1. 

CSR reports have no effect on customer-based measures of financial performance.

2.2. Moderators of the Impact of CSR Reports

2.2.1. Assuring the Quality of CSR Reports

The previous literature on socially responsible firms and the halo effect finds that customers have favorable attitudes toward the products from these firms. However, despite the favorable perceptions of socially responsible and sustainable firms, there are three issues regarding the actual use of CSR reports by customers. First, there are no quality standards for data. Second, CSR reports are prepared for all stakeholders whose expectations and interests are diverse and therefore conflict with each other. Third, much of the data in CSR reports is qualitative in nature, requiring special skills to understand and to interpret [55]. The complex and technical information contained in CSR reports causes low readership of these reports [56]. This suggests that there is a potential that customers will be skeptical of the information contained in a firm’s socially responsible and sustainable report [57]. Therefore, customers may look for some indicator of the quality of the information contained in CSR reports.

There are various ways to improve the perceived quality of a firm’s disclosures. One of the arguments in favor of IFRS is that the use of standardized formats for financial information improves its quality, which benefits investors [58]. As with the arguments in favor of IFRS, there is also recognition that the standardization of sustainability disclosures would improve their quality [59]. While there is not a requirement for standardized formats for CSR and sustainability disclosures, many firms have adopted the GRI frameworks [60,61].

The Global Reporting Initiative (GRI) [31,62,63] has established a standardized framework for firms that describes the types of information to be included in their socially responsible disclosures. The GRI framework offers a high degree of consistency in sustainability disclosures and therefore has gained acceptance as a viable approach to CSR reporting. Establishing an accepted framework should improve customers’ acceptance of the firm’s social responsibility disclosures, provide evidence that sustainability is part of the firm’s business model, and finally reduce information asymmetry [27]. While CSR reports have been shown to be of some value, there have been arguments that, in addition to using GRI frameworks, there are other ways to improve the value of these reports [64].

The GRI frameworks have gone through a transition in terms of the items to be included if firms want to indicate that their report conforms to GRI standards. In 1999, there were guidelines that suggested firms make certain sustainability information available if they wanted to indicate that their disclosure complies with GRI guidelines [31]. In 2006, the GRI released the G3 Guidelines, which included application levels [62]. These application levels were self-declared and ranged from C to A, which included specific categories of the GRI criteria that should be included in the report. The quality of the reports increased (for levels B and A) with the inclusion of additional criteria, management disclosures concerning the criteria, and at level A, a segment supplement. Beginning in 2013, reports filed using the GRI framework need to comply with the new G4 guidelines. The primary difference between the G4 guidelines and previous GRI guidelines is the inclusion of materiality standards for the content of the reports [65]. Each of these transitions not only increases the coverage of a GRI report but also allows stakeholders to rely on the consistency and improves the quality of GRI reports [66]. However, as the application levels are self-declared and the data included in the report is voluntary, some level of external assurance is also desirable.

Each of the GRI’s G3 application levels (A, B, & C) can have a “+” added to the level to indicate that the report has been verified [62]. There is not a specific requirement as to the type of firm performing the verification; however, the expectation is that the provider will be independent from the firm. The GRI has identified three types of entities that can perform assurance of their reports, as each has a different type of expertise that relates to different aspects of the sustainability report. Accounting firms have a global presence and are the predominant provider of these services (64% in 2012), as they have a natural competency in assurance services and are well versed in providing financial and non-financial assurance. Sustainability services firms have been the next largest provider of assurance services for GRI reports (23% in 2012). While they certainly do not have the global reach of audit firms, they do have expertise in many of the issues disclosed in the reports and are also experienced in specific stakeholder issues. The third type of assurance provider is engineering firms (13% in 2012). They have the necessary technical expertise to review complex processes [32] (the statistics come from the GRI website [67]). Adding these external assurances provides stakeholders with a degree of confidence that the information contained in a GRI report has some level of quality, but there is also a potential problem of selection bias.

Firms can choose to create the content and whether to acquire assurance for GRI reports. Firms choose to pay to have separate assurance on these reports [68,69,70]. These CSR disclosures that have been assured are associated with enhanced financial performance [71,72,73,74]. However, the value of external assurance of CSR reports (or any non-financial disclosure) may not outweigh its cost [75]. Therefore, the potential exists for firms to select the type of assurance report that is most favorable to their sustainability practices and to only seek an assurance engagement that is most beneficial [28]. Firms that make socially responsible disclosures achieve greater trust and stature among customers [76,77,78]. For these reasons, it may be particularly important for firms to provide customers with some form of assurance as a more objective measure of their CSR report’s quality. This suggests that the degree of coverage and some type of external assurance will impact customers’ appreciation of the content of a CSR report [56,79,80]. The above discussion leads to the following hypotheses:

H2. 

The degree of coverage of CSR reports has an impact on customer-based profitability measures.

H3. 

External assurance positively moderates the effect of CSR reports on customer-based profitability.

2.2.2. The Impact of Advertising on CSR Reports

While arguments have been made that customers should be aware of a firm’s CSR actions, it is argued that these reports are too complex and technical for ordinary customers to fully appreciate their content [55]. While the quality of a CSR report has been shown to affect customers’ views of firms, the firms can also choose other approaches to disclose their CSR and sustainable actions. In the previous section, the importance of external assurance on CSR was discussed; however, firms can also use advertising as an approach to provide customers with information about their socially responsible actions and the impact of these actions on their products.

Chang, Tsai, & Huang [81] conclude that different environmental advertising strategies can impact customers’ attitudes toward firms. Their study shows that these attitudes are positively associated with purchase intentions. Veche [82] documents that CSR representations in slogans can impact customers’ views of the sustainability practices of banks. Erdem & Swait [83] examine whether firms can reduce uncertainty of a product’s characteristics by providing certain signals to customers. Depending on the type of signal, these signals can provide credibility to certain characteristics of the brand and improve the customers’ information about firms’ products [84]. Evidence indicates the potential that positive CSR messages will support the functional aspects of products and can impact customers’ perceived quality of products, as well as the price they will be willing to pay [85,86]. This supports Porter’s [87] identification of marketing as a primary activity in creating firm value. Krasnikov & Jayachandran’s [88] study reinforces Porter’s arguments as they find that marketing capability has a stronger impact on firm performance than research and development or operational capabilities. Thus, marketing may be an important factor in providing customers with information about firms’ socially responsible and sustainable actions and eliciting their support [89]. As a result, firms use advertising to communicate their CSR actions [90]. There is some evidence that customers see advertising of CSR actions as impacting on the firm’s reputation and thus their support for the firm [91]. Therefore, there is reason to support the use of advertising as a tool to enhance customers’ view of the legitimacy of the firm’s sustainable and CSR actions [29,30]. Guenther & Guenther [92] document that advertising can drive increased sales, particularly in the B2B segment. Further, a history of socially responsible advertising can also divert attention from any of the firm’s socially irresponsible actions [90]. This makes advertising a possible approach to increase customers’ awareness of firms’ CSR actions [93]. Thus, Assaf, Josiassen, Ahn, & Mattila’s [94] argument that there is a link between advertising, CSR, and a firm’s financial performance is plausible. However, as with other disclosures of CSR actions, the firm’s discretion in the types of advertising messages has yielded inconsistent results on the firm’s performance [56,79]. These arguments suggest that in addition to other approaches a firm might use to increase the credibility of their CSR disclosures, advertising also has the potential to increase the value of these messages and leads to the following hypothesis.

H4. 

Advertising positively moderates the effect of CSR reports on customer-based profitability.

3. Methodology

3.1. Data and the Econometric Model

The study uses financial data obtained from Compustat North America and non-financial data about CSR reports and external assurance from the GRI website [95], as well as firms’ annual reports. The sample of publicly owned firms was obtained from the GRI reporting registry. This registry consists of 4574 firms worldwide that publish their CSR reports in accordance with GRI guidelines. These firms have different legal statuses, such as public, private, and state-owned firms, cooperatives, and not-for-profit entities with different application levels (A+, A, B+, B, C+, and C, where “+” denotes third-party assurance). The sample resulting from this registry includes 465 publicly owned North American (US and Canadian) firms. Since different customer segments may exhibit variations in their CSR commitments, we make a distinction between B2B and B2C firms in the sample [95,96]. Consistent with prior research, to test our hypotheses, we adopt the following models by using cross-sectional OLS and RR estimators [93,96].

prof_it = α₀ + α₁ csr_it + α₂ exas_it + α₃ csr_it × exas_it + α₄ adv_it + α₅ controls _it + ind _it + year _it + ε_it

(1)

prof_it = α₀ + α₁ csr_it + α₂ adv_it + α₃ csr_it × adv_it + α₄ controls_it + ind _it + year _it + ε_it

(2)

3.2. Dependent Variables

In this study, we use both gross profit scaled by total assets and sales growth as dependent variables (prof) [97,98]. It is argued that gross profit is the cleanest accounting measure of true economic profitability, and the profitability becomes more polluted and less related to true economic profitability as one goes farther down the income statement [98]. This profitability metric is also particularly dependent on customers’ reactions to messages from the firm (CSR reports and advertising) and represents customers’ willingness to pay higher prices and is closely associated with profitability arising solely from sales. Sales growth is also a widely used measure of profitability and is also a direct result of increased customer loyalty when compared to sales growth at firms that do not have the same level of CSR activities. These two measures, gross profit and sales growth, can be used to estimate a firm’s potential to survive in highly competitive markets [99].

3.3. Independent and Control Variables

Our independent variables include csr, exas, and adv. csr, which is a dummy variable that indicates whether a company produces a CSR report. exas refers to a dummy variable for external assurance on CSR reports by independent assurance providers. adv is a proxy for customer awareness about the company and refers to advertising intensity [96,100,101]. Consistent with prior research [102,103], we use the following control variables: capital intensity (capin) is computed as total assets/number of employees; efficiency (eff) is computed as COGS/sales; leverage (lev) is total assets/total debt; the current ratio (currt); and company size (size) as total assets. industry and year dummies are also included in regressions. In addition, we use the independent auditor’s review of internal controls (auopic), as quality internal controls have been shown to also impact the quality of non-financial business processes [37]. (For a detailed description of the variables used in this study, see Appendix A).

4. Results of the Empirical Analyses

4.1. Univariate Analysis

Table 1. (A) Descriptive statistics—overall sample, (B) descriptive statistics—B2B sector, (C) descriptive statistics—B2C sector.

	Mean	SD	Q1	Median	Q3
A
gp	0.387	0.223	0.231	0.339	0.504
lngp	0.315	0.154	0.209	0.292	0.408
grwsale	0.044	0.166	−0.016	0.031	0.089
csrrep	0.814	0.389	1	1	1
exas	0.218	0.413	0	0	0
auopic	0.976	0.152	1	1	1
capin	848.249	1387.168	226.412	467.722	942.147
lncapin	6.130	1.093	5.422	6.148	6.848
size	33,443.83	51,682	5098.9	10,912	33,749
lnsize	9.485	1.363	8.537	9.298	10.427
lev	0.302	0.201	0.172	0.283	0.404
lnlev	0.254	0.139	0.158	0.249	0.339
adv	0.029	0.037	0.006	0.016	0.039
lnadv	0.028	0.034	0.006	0.016	0.038
B
gp	0.344	0.147	0.249	0.326	0.426
lngp	0.289	0.109	0.222	0.282	0.355
grwsale	0.046	0.165	−0.02	0.028	0.090
csrrep	0.806	0.396	1	1	1
exas	0.309	0.463	0	0	1
auopic	0.978	0.149	1	1	1
capin	1022.949	1300.154	352.474	589.940	1072.776
lncapin	6.431	1.004	5.865	6.380	6.978
size	27,823.46	40,192.69	4412.563	9020.393	27,816.3
lnsize	9.344	1.326	8.392	9.107	10.234
lev	0.296	0.168	0.182	0.260	0.409
lnlev	0.251	0.127	0.167	0.231	0.343
adv	0.016	0.020	0.003	0.007	0.020
lnadv	0.016	0.019	0.003	0.007	0.020
C
gp	0.412	0.253	0.221	0.350	0.561
lngp	0.330	0.172	0.200	0.300	0.445
grwsale	0.042	0.168	−0.016	0.033	0.084
csrrep	0.819	0.390	1	1	1
exas	0.167	0.374	0	0	0
auopic	0.976	0.154	1	1	1
capin	751.513	1424.499	171.763	398.044	870.009
lncapin	5.963	1.105	5.146	5.987	6.769
size	36,555.98	56,837.68	5496.8	12,466	35.387
lnsize	9.564	1.377	8.612	9.429	10.474
lev	0.306	0.217	0.169	0.292	0.403
lnlev	0.255	0.145	0.156	0.255	0.339
adv	0.036	0.041	0.009	0.022	0.045
lnadv	0.034	0.038	0.009	0.021	0.044

reports descriptive statistics for the overall sample and the B2B and B2C segments. Except for csr and exas, natural logarithm values are reported. For the overall sample, gp has a mean of 0.387. The mean gp in B2C is higher than that in the B2B segment (0.412 vs. 0.344). This difference could be due to firms in the B2B segment purchasing larger quantities, resulting in higher bargaining power and effectively negotiating prices and payment terms [50]. The mean grwsale of B2B is higher than that of the B2C segment (0.046 vs. 0.042), suggesting that the B2B segment has a greater sales growth as compared with firms in the B2C segment. The mean of csr is 0.814, so over 80% of firms have some sort of CSR report, but less than 22% (exas is 0.218) seek external assurance for their CSR reports. While the percentage publishing reports is similar, B2B CSR reports have more external assurance than those in the B2C segment (0.309 vs. 0.167), as suggested by Han & Lee [47]. The mean of adv is 0.029 for the overall sample, while the B2C segment has a higher mean of adv than that for the B2B segment (0.036 vs. 0.016), suggesting that firms in B2C spend more on advertising to promote the value of their brands [93,104], and that B2B firms see advertising as fulfilling a secondary function [91]. (The CMO survey has found that in 2023 B2C marketing budgets tend to be higher than B2B budgets, accounting for 13.9% and 9.3% of sales revenues, respectively [105]). Finally, the analysis of the Spearman and Pearson correlations suggests that there are no multicollinearity issues.

4.2. The Results of the Multivariate Analysis

When we conduct regression diagnostics after OLS estimations, we find that gp, capin, and size are not normally distributed. This suggests that OLS estimations might be influenced by outliers. So, we employ RR in addition to OLS to examine the sensitivity of our results [106]. Gross profit (gp) is the dependent variable in

Table 2. Cross-section regression results (dependent variable is gross profit).

	Reg. 1	Reg. 2	Reg. 3	Reg. 4	Reg. 5	Reg. 6
	OLS	RR	OLS	RR	OLS	RR
	Overall	Overall	B2B	B2B	B2C	B2C
csr	0.020 **	0.019 **	0.017	0.003	0.021 *	0.020 *
	(0.009)	(0.009)	(0.013)	(0.012)	(0.011)	(0.011)
exas	−0.053 ***	−0.052	−0.010	−0.024	−0.147 ***	−0.142
	(0.020)	(0.037)	(0.018)	(0.030)	(0.019)	(0.130)
csr × exas	0.050 **	0.062 *	0.059 ***	0.074 **	0.104 ***	0.106
	(0.022)	(0.037)	(0.021)	(0.031)	(0.020)	(0.131)
auopic	0.059 ***	0.068 ***	0.069 ***	0.065 **	0.058 **	0.069 ***
	(0.022)	(0.021)	(0.020)	(0.028)	(0.028)	(0.026)
capin	−0.063 ***	−0.067 ***	−0.020 ***	−0.018 ***	−0.082 ***	−0.088 ***
	(0.004)	(0.003)	(0.006)	(0.005)	(0.004)	(0.004)
size	−0.014 ***	−0.013 ***	−0.002	−0.003	−0.021 ***	−0.022 ***
	(0.003)	(0.003)	(0.004)	(0.004)	(0.003)	(0.003)
lev	−0.068 ***	−0.079 ***	−0.069 *	−0.100 ***	−0.089 ***	−0.107 ***
	(0.025)	(0.024)	(0.039)	(0.035)	(0.029)	(0.029)
adv	1.084 ***	1.097 ***	1.348 ***	1.381 ***	1.098 ***	1.153 ***
	(0.098)	(0.096)	(0.180)	(0.214)	(0.104)	(0.109)
constant	0.761 ***	0.770 ***	0.361 ***	0.374 ***	0.944 ***	0.971 ***
	(0.039)	(0.034)	(0.043)	(0.044)	(0.047)	(0.041)
Observations	1605	1605	572	572	1033	1033
R2	0.337		0.174		0.472
R2-adj.	0.329		0.147		0.462
Prob > F		0.000		0.000		0.000

Notes: Robust standard errors are in parentheses [*** p < 0.01, ** p < 0.05, * p < 0.1]. B2B and B2C represent the business-to-business and business-to-customer sectors. Dependent variable is gross profit scaled by total assets (gp). csr is a dummy variable for CSR report. exas is external assurance on CSR reports provided by third-party assurance providers. auopic is a dummy variable for independent audit on internal control. capin is capital intensity. size refers to firm size proxied by total assets. lev refers to leverage. adv is advertising expenditure. industry and year dummies are included in regressions.

, and the results allow us to conclude that firms that publish CSR reports (csr) have gross profits (gp) that are significantly positive (Reg. 1, 2, 5, and 6). This result leads to the rejection of the null hypothesis H1 and indicates that CSR reports have a significant and positive impact on gross profit for the overall sample and for the B2C segment, but not in the B2B segment. The lack of a significant relationship between B2B firms and gross profit supports Hoejmose et al.’s [103] argument that personal relationships and trust play an important role for firms to improve their performance and competitive position in the B2B segment rather than focusing on CSR activities. There is a significant and negative association between exas and gp for the entire sample and B2C (Reg. 1 and 5). However, we find positive and significant coefficients on csr × exas in all regressions (except 6), suggesting that CSR reports supported by external assurance increase profitability for the entire sample and the B2B and B2C segments, and thus H3 is supported (coef. = 0.051, p < 0.05; coef. = 0.062, p < 0.10; coef. = 0.059, p < 0.05; coef. = 0.074, p < 0.05; coef. = 0.104, p < 0.01).

In addition to firms in the B2C segment, firms in the B2B segment increasingly demand that their suppliers engage in CSR activities. These firms, such as IBM, Sony, and Philips, require their suppliers to comply with these standards in order to continue their business relationships [47]. Therefore, suppliers should take into account CSR activities for maintaining and improving their business relations with their business partners [47,107].

Responses to questionnaires suggest that customers are willing to pay more and will continue to purchase products from socially responsible firms [85,108]. If firms produce eco-friendly goods, they will enjoy the halo effect suggested by Madden et al. [21], so that the public will have a positive view of the firm’s products even if they have never tried their products before. However, to obtain these benefits, firms must also make customers aware of their CSR activities. One way to make customers aware of these is to obtain external assurance, which would also increase the credibility of these reports [51]. Our results for all regressions in Table 2 (except 6) support this.

Another approach that firms use to increase customers’ awareness and thereby their profitability is through advertising [93]. In

Table 3. Cross-section regression results (gross profit is dependent variable).

	Reg. 1	Reg. 2	Reg. 3	Reg. 4	Reg. 5	Reg. 6
	OLS	RR	OLS	RR	OLS	RR
	Overall	Overall	B2B	B2B	B2C	B2C
csr	0.003	0.008	0.008	0.006	−0.025	−0.024
	(0.016)	(0.014)	(0.019)	(0.016)	(0.022)	(0.020)
adv	−0.028	−0.019	−1.411	0.427	−1.309 **	−1.506 ***
	(0.469)	(0.399)	(1.600)	(0.918)	(0.509)	(0.493)
csr × adv	0.807 *	0.843 **	2.701 *	0.852	1.459 ***	1.620 ***
	(0.489)	(0.414)	(1.611)	(0.939)	(0.523)	(0.505)
auopic	0.061 **	0.057 **	0.103 ***	0.080 **	0.049	0.048
	(0.030)	(0.026)	(0.034)	(0.035)	(0.033)	(0.033)
capin	−0.128 ***	−0.118 ***	−0.067 ***	−0.056 ***	−0.138 ***	−0.135 ***
	(0.005)	(0.004)	(0.007)	(0.006)	(0.005)	(0.006)
size	−0.019 ***	−0.016 ***	−0.015 ***	−0.016 ***	−0.026 ***	−0.024 ***
	(0.003)	(0.003)	(0.004)	(0.004)	(0.004)	(0.004)
lev	−0.199 ***	−0.203 ***	−0.219 ***	−0.232 ***	−0.226 ***	−0.232 ***
	(0.033)	(0.031)	(0.048)	(0.043)	(0.039)	(0.039)
eff	−0.618 ***	−0.559 ***	−0.394 ***	−0.354 ***	−0.882 ***	−0.866 ***
	(0.035)	(0.032)	(0.043)	(0.037)	(0.062)	(0.056)
grw	−0.212 ***	−0.228 ***	−0.286 ***	−0.288 ***	−0.174 ***	−0.181 ***
	(0.023)	(0.020)	(0.039)	(0.029)	(0.025)	(0.025)
constant	1.666 ***	1.555 ***	1.087 ***	1.033 ***	1.995 ***	1.957 ***
	(0.055)	(0.047)	(0.069)	(0.064)	(0.063)	(0.062)
Observations	1401	1401	500	500	901	901
R2	0.597	0.573	0.475	0.493	0.672	0.641
R2-adj.	0.592	0.567	0.455	0.474	0.665	0.634

Notes: Robust SE in parentheses [*** p < 0.01, ** p < 0.05, * p < 0.1]. B2B and B2C represent the business-to-business and business-to-customer sectors. Dependent variable is gross profit scaled by total assets (gp). csr is a dummy variable for the CSR report. adv is advertising intensity. auopic is a dummy for independent audit on internal control. capin is capital intensity. size refers to firm size scaled by total assets. lev refers to leverage. eff refers to efficiency. industry and year dummies are included in regressions.

, gross profit (gp) is used as the dependent variable, and csr and adv are used as independent variables; csr becomes insignificant in all regressions. Although the relationship between adv and gp is negative in all regressions, it is only significant in Reg. 5 and 6 for the B2C segment (coef. = −1.309, p < 0.05; coef. = −1.506, p < 0.01). Normally, a positive impact of advertising is expected on gross profit. This negative association may be due to customers’ inclination to be skeptical of advertising claims [109,110]. Even if firms could raise profits in the short term by false advertising, they would lose their market share and incur financial losses in the medium term by being required to pay higher penalties [111].

The positive and significant coefficients on csr × adv for the overall sample and in the B2B and B2C segments indicate that advertising intensity can enhance the credibility of CSR reports and the credibility of firms’ socially responsible actions, which increases the firms’ profitability. This result offers support for our H4. To put it differently, advertising expenditures increase the impact of CSR reports on profitability, because advertising helps customers (in both B2C and B2B segments) to become aware of firms’ CSR activities [93,94]. It is argued that advertising that provides customers with a high level of awareness about CSR activities and superior CSR performance is likely to contribute to brand value and legitimacy of firms, thereby increasing the demand for firms’ products [112].

4.3. Robustness Tests and Additional Analysis

In this section, we report the results of several additional tests conducted to check the robustness of our findings. In cross-sectional datasets, heteroskedasticity could be an issue, and therefore feasible generalized least squares (FGLS) is a suggested alternative to OLS [113,114]. When we replicated our estimations using FGLS, our findings confirmed the previously reported OLS results.

Table 4. Cross-section regression results—additional analysis (sales growth is the dependent variable).

	Reg.1	Reg.2	Reg.3	Reg.4
	OLS	RR	OLS	RR
			G4-Core	G4-Core
csr	−0.521 ***	−0.631 ***	−0.520 ***	−0.637 ***
	(0.039)	(0.113)	(0.051)	(0.115)
acct	−0.560 ***	−0.720 ***	−0.523 ***	−0.728 ***
	(0.053)	(0.117)	(0.072)	(0.123)
csr × acct	0.542 ***	0.689 ***	0.510 ***	0.682 ***
	(0.051)	(0.118)	(0.068)	(0.124)
auopic	0.065 *	0.039	−0.003	−0.007
	(0.038)	(0.029)	(0.044)	(0.040)
size	−0.001	0.003	−0.006	0.003
	(0.008)	(0.004)	(0.011)	(0.005)
eff	−0.107	−0.101 ***	−0.079	−0.071 **
	(0.068)	(0.027)	(0.102)	(0.035)
capin	0.009	−0.009 **	0.017	−0.011 **
	(0.009)	(0.004)	(0.012)	(0.005)
constant	0.471 ***	0.684 ***	0.608 ***	0.845 ***
	(0.131)	(0.125)	(0.207)	(0.178)
Observations	615	615	397	397
R2	0.171	0.259	0.165	0.273
R2-adj.	0.148	0.239	0.132	0.244

Notes: Robust standard errors in parentheses [*** p < 0.01, ** p < 0.05, * p < 0.1]. G4-Core represents “G4, In accordance—Core” by GRI. The dependent variable is the natural logarithm of the growth of sales (grwsale). csr is a dummy variable for the CSR report. acct is a dummy variable for external assurance provided by accounting firms. auopic is a dummy variable for external audit on the effectiveness of internal control. size refers to company size scaled by total assets. eff refers to efficiency. capin is capital intensity. industry and year dummies are included in regressions.

presents the results of regressions of csr and exas when the external assurance is provided by accounting firms (acct) and the dependent variable is sales growth (grwsale). We find that csr × acct is significantly positive in all regressions, suggesting that the assurance provided by accounting firms reverses/reduces the negative effect of csr on grwsale. Reg. 3 and 4 are for CSR reports that are prepared in compliance with “G4-Core” (“An organization can choose between two options to meet G4’s ‘in accordance’ criterion. The ‘Core’ option contains the essential elements of a sustainability report, while the ‘Comprehensive’ option builds on the ‘Core’ option but requires additional information on governance” [114]). It is argued that preparing CSR reports using G4-Core formats results in higher quality reports in accordance with G4-Core. The results show that when CSR reports are prepared using the G4-Core and accounting firms provide external assurance, this leads to an increase in sales growth. (The materiality is emphasized throughout the G4 Guidelines [94]). This is an important finding, as preparing CSR reports using the G4 framework requires additional effort by the firm, as it must consider the broader boundary of their operations and so incorporate a review of their impact on the supply chain for their products [65]. Our result indicates that this additional effort is worthwhile, as it increases our measure of customer loyalty, i.e., sales growth.

Table 5. Cross-section regression results—additional analysis (dependent variable is the gross profit).

	Reg.1	Reg.2	Reg.3	Reg.4	Reg.5	Reg.6
	Overall	B2B	B2C	Overall	B2B	B2C
csr	0.006	−0.059	0.092 ***	0.020	0.088 ***	−0.045
	(0.032)	(0.049)	(0.026)	(0.024)	(0.012)	(0.050)
exas	−0.074 *	−0.100 *	−0.045 *
	(0.043)	(0.057)	(0.026)
adv				−3.874 *	4.475 ***	−8.779 **
				(2.065)	(1.209)	(4.219)
csr × exas	0.062	0.129 **	−0.030
	(0.044)	(0.058)	(0.030)
csr × adv				4.415 **	−3.270 ***	9.003 **
				(2.079)	(1.189)	(4.276)
auopic	0.062 ***	0.043 **	0.083 ***	0.088 ***	0.063 ***	0.094 ***
	(0.018)	(0.017)	(0.026)	(0.031)	(0.018)	(0.036)
capin	−0.073 ***	−0.055 ***	−0.087 ***	−0.100 ***	−0.028 ***	−0.128 ***
	(0.004)	(0.004)	(0.008)	(0.007)	(0.009)	(0.008)
size	−0.003	−0.005	−0.004	−0.001	−0.016 ***	0.010 *
	(0.003)	(0.004)	(0.005)	(0.004)	(0.005)	(0.005)
eff	−0.608 ***	−0.561 ***	−0.646 ***	−0.556 ***	−0.319 ***	−0.688 ***
	(0.053)	(0.068)	(0.052)	(0.044)	(0.046)	(0.062)
lev	−0.036	−0.022	−0.086 **	−0.056 *	−0.150 ***	−0.057
	(0.030)	(0.041)	(0.037)	(0.032)	(0.048)	(0.038)
curr	−0.007	0.007	0.006	−0.011	−0.003	0.006
	(0.008)	(0.009)	(0.011)	(0.010)	(0.010)	(0.013)
constant	0.967 ***	0.885 ***	1.053 ***	1.076 ***	0.625 ***	1.267 ***
	(0.073)	(0.096)	(0.082)	(0.081)	(0.083)	(0.075)
Observations	1161	749	412	489	167	322
R-squared	0.559	0.633	0.541	0.491	0.369	0.577
R2-adj.	0.555	0.628	0.531	0.481	0.333	0.565

Notes: Robust SE in parentheses [*** p < 0.01, ** p < 0.05, * p < 0.1]. B2B and B2C represent business-to-business and business-to-customer sectors, G4-Core represents “G4, In accordance—Core” by GRI. Dependent variable is natural gross profit scaled by total assets (gp). csr is a dummy variable for the CSR report. exas is external assurance on CSR reports provided by third-party assurance providers. adv is advertising intensity. auopic is a dummy for external audit on the effectiveness of internal control. capin is capital intensity. size refers to firm size scaled by total assets. eff refers to efficiency. lev is leverage. curr is the current ratio. industry and year are included in regressions.

reports the regression results only for companies that publish their CSR reports in compliance with “G4-Core” where gross profit is the dependent variable. Table 5 reports all the independent variables. In Reg. 3, we find that csr is significantly and positively associated with gp for firms in the B2C segment that report CSR activities in accordance with G4-Core Guidelines (coef. = 0.092, p < 0.01). So, comprehensive CSR reports increase gross profits in this segment. On the other hand, the positive association between csr × exas on gp suggests that CSR reports with third-party assurance enhance gross profits (coef. = 0.129, p < 0.05). In Reg. 5, we find a positive association between csr and gp. Thus, it appears that comprehensive CSR reports contribute to gross profit in the B2B segment (coef. = 0.088, p < 0.01). In this segment, although there is a positive relationship between adv and gp, the significant and negative coefficient on csr × adv suggests that CSR reports in companies with high advertising expenditures gross profits. In contrast, advertising appears to enhance the impact of CSR reports on the B2C segment. The positive and significant coefficient on csr × adv in Reg. 4 and 6 indicates the importance of advertising in firms that publish their CSR reports in accordance with G4-Core (overall sample and in the B2C segment).

Overall, these results suggest that, while G4-Core CSR reports alone may have a mixed impact on financial performance, the addition of external assurance makes their impact highly significant. This may be due to the CSR report’s highly complex and technical language. Therefore, the addition of either assurance on the report itself or the communication of its content, through advertising, seems to be of value to customers, as they show by paying higher prices and/or increasing their purchases. To put it differently, retail customers exert more pressure on firms to adopt environmental protection systems and practices than commercial and industrial customers [115,116,117].

5. Discussion, Limitations, and Policy Recommendations

The objective of this study is to examine the relationship of CSR reports with customer-based measures of financial performance for North American firms. We also examine how different types of assurance and advertising expenditures affect this relationship. The study makes a noteworthy contribution, as it demonstrates that while customers should be more willing to support socially responsible firms, their support is contingent on some type of external assurance of the report. Our results support the findings of other researchers, who argue that the technical and complex nature of CSR disclosures may inhibit customers from fully appreciating the content of these reports, and therefore, it is important for CSR reports to have some type of assurance provided by an external party. The study examines direct assurance provided by accounting, engineering, and consulting firms. The results show that any form of external assurance enhances customer-based measures of financial performance. We also find that the firm’s use of advertising has a significant impact on the value customers place on CSR disclosures. Furthermore, firms’ measures of customer-based financial performance are enhanced when their CSR disclosures are made using the GRI (G4) framework. Finally, we also confirm that there are differences between customers in the B2B and B2C segments. B2B customers are interested in the external assurance of CSR reports, while B2C customers respond to CSR actions when the firm advertises their actions. These results are consistent with the findings of prior research of the B2B and B2C segments.

The findings of this study are important for public and private regulatory organizations, such as the Security and Exchange Commission, the Financial Accounting Standards Board, the International Accounting Standards Board, and GRI. These organizations are continually expected to increase their efforts to issue standards about firms’ CSR actions. The results suggest that certain policy recommendations are appropriate. First, while it is important for firms to be socially responsible, it is also important to provide customers with some type of assurance. While the assurance of CSR reports is certainly not performed according to standards in place of financial statements, our results show the assurance is important. Therefore, continued work by GRI and ISSB to establish standards for CSR and their assurance is critical to ensure information about sustainability has the same quality as financial statements. A second policy implication of our study is that customers place value on information contained in advertising messages. Customers (those in the B2B segment) do not place value on advertising messages; however, those in the B2C segment are influenced by advertising. Therefore, evaluating the quality of CSR messages contained in advertising should be reviewed.

As with all studies, there are limitations to this study. First, CSR reports and GRI standards are not mandatory. The voluntary nature of GRI standards causes firms to selectively disclose sustainability information, which may mislead stakeholders’ perception towards firms’ risk and performance and reduce the efficiency of GRI standards. Second, many firms allocate inadequate resources to CSR reports and the training programs for their staff to acquire the required level of skills. Third, while customers place some value on advertising messages, the accuracy of these messages could be questioned. There is evidence that some firms use false advertising to overstate the value of their products by misleading their descriptions. Therefore, additional research could validate whether customers see these messages as providing accurate information. False advertising would not only harm customers but also cause firms to lose their loyal customers, thereby affecting financial performance adversely. An additional limitation concerns the restricted geographic nature of the results. While North America is a large market, future studies could examine whether the results are consistent across other geographic regions.