How Do Corporate Social Responsibility Engagements Drive Consumer–Company Identification in Singapore?

Abstract

Companies expend significant financial resources on corporate social responsibility (CSR) activities to enhance their image. This study had two objectives. By drawing on three fundamental consumer values as moderators, the first aim was to discover how a company’s CSR engagement influences consumer–company identification (CCI) and consumers’ purchase intention for its products and services. The second was to uncover the type of consumer likely to identify with a company engaging in CSR activities. This study presents an exploratory analysis of social media postings by eight companies. An empirical study is conducted using partial least squares structural equation modelling (PLS-SEM) based on survey data from 217 questionnaires collected in Singapore. This study is one of the first to consider what type of consumer would likely identify with a company performing CSR activities. Results show that local-community-focused CSR tends to influence the CCI of egoistic consumers, while environment-focused CSR activities, such as the creation of environmentally-friendly products, drive biospheric consumers’ identification. Broader humanity-focused CSR, such as fair-trade initiatives, significantly influences the CCI of egoistic consumers, but not of altruistic consumers. These findings demonstrate to other economies the significance of local and global companies’ CSR practices and how such activities should be aligned to the CCI of their customer base unique to the region.

1. Introduction

Corporate social responsibility (CSR) is defined as a context-specific, strategic, proactive, and synergistic business philosophy, while attention is paid to broader economic, environmental, and social issues in a balanced way [1]. In Accenture Strategy’s global survey of almost 30,000 consumers, 62% say they want companies to take a stand on sustainability, transparency, and fair employment to align with their own beliefs. Nearly half (47%) of consumers will walk away if they are disappointed with a brand’s words or actions on a social issue [2]. As a result, more and more companies are committing resources to CSR and communicating it to their customers. However, although many companies embrace this broad vision of CSR, it is not uncommon for them to be hampered by poor coordination and a lack of logical connections between their various programmes [3] and to fail to establish a stand that aligns with their customers’ expectations. Numerous studies have reported mixed empirical evidence for the impact of CSR on a company’s performance [4,5,6,7]. Because companies and business practitioners expend enormous resources on CSR activities that ultimately affect the bottom line and product costs, understanding this aspect is crucial to business operations.

In tackling the mixed findings, several scholars have used consumer–company identification (CCI) to understand the underlying mechanism through which CSR affects consumer reactions. CCI is the degree to which customers see a business as equivalent to themselves [8]. A strong consumer–company relationship arises when customers equate their identity with the corporate identity [9]. Therefore, a stronger CCI may result in a higher purchase intention for the company’s products and services [10], leading to a better bottom line for the company.

Studies that link CSR with CCI fall into two fields. The first strand shows CCI as a mediator between a company’s CSR engagements and various consumer reactions. For example, Huang, Cheng and Chen [11] indicated that the restaurants’ service quality and CSR engagement enhance CCI, positively influencing customer loyalty. Similarly, Raza et al. [12] reported the mediating role of CCI between CSR and customer loyalty in the banking industry. The second strand questions the universality of such a mediating relationship and explores the conditions or contingencies that enable CSR to help CCI. For example, Lii and Lee [13] looked into different types of CSR and found that CSR in the forms of philanthropy, cause-related marketing, and sponsorship has a significant effect on CCI. Li, Chen and Qing [10] revealed that the impact of CSR perception on CCI is contingent on the type of CSR, i.e., whether they are must-be CSR (e.g., meet quality standards and treat employees equally) or one-dimensional CSRs (e.g., set decent working conditions and prevent waste). These conditions are mainly the strategies of a company’s choice. However, consumers, as a crucial group of stakeholders that the CSRs target to impress, are left out of the picture. Consumers are not homogeneous. They differ in terms of their values and concerns [14], and hence have different expectations and perceptions of a company’s CSRs [15,16].

This study, thus, tries to fill up the gap by answering the question: what types of consumers are more responsive to different CSR practices and are more likely to formulate a closer CCI and higher purchase intention? We introduce a typology of CSR engagements based on the work by Carrol [17] and categorize CSR activities into three types: local-community-focused CSR, environment-focused CSR, and broader humanity-focused CSR. We also categorize consumer values into three types: egoistic, biospheric, and altruistic, following Verma, Chandra and Kumar [18]. Drawing on stakeholder theory and social identity theory, we examine how consumers with different values formulate CCI with a company that engages in various CSR activities.

Empirically, our questionnaire survey is based on 217 consumers of eight companies selected from Singapore. Singapore provides an ideal testing ground since it is home to well-established local and global corporations [19]. Many of them have shown a great appreciation of CSR and engage in various CSR activities [20,21]. At the same time, Singapore, a representative of South-East Asia and the tropical region, also possesses a multicultural setting that accommodates pluralistic consumer types with diverse values [22].

Our results from partial least squares structural equation modelling (PLS-SEM) analysis have shown that not every type of CSR can help the company create CCI with every customer. Instead, an alignment between the type of CSR and consumer values is the precondition of a strong CCI, further leading to a higher purchase intention. Such findings bring new insights to the literature on consumer responses to CSR, both theoretically and practically. Theoretically, the results enrich the CSR literature by showing a refined mediation mechanism, through which a company’s CSR influences consumer reaction, i.e., a mediation relationship between CSR, CCI, and purchase intention moderated by consumer values. This study also provides several practical implications, which matter for the sustainable development of tropical regions. The tropical zone is the most biologically diverse region on Earth, hosting about 80% of the planet’s terrestrial species and over 95% of its coral species and mangroves. It has most of the world’s cultural and linguistic diversity, with extremely high language diversity in Melanesia, the Amazon, equatorial Africa, South-East Asia and India. It is growing about 20% faster than the rest of the world, yet with varying levels of economic development [23]. Through CSR engagements, large and small companies are essential contributors to the region’s sustainable development. Our findings inform companies about targeting CSR engagements that benefit the environment and society, while improving their bottom line, which is particularly important for conducting business in the tropical region.

The remainder of the paper is organised as follows: Section 2 presents a literature review of the effect of CSR on firm performance and our four research hypotheses. Section 3 describes our data and introduces established measures of CSR, CCI, and consumer values. Our main results related to the influence of different types of CSR and values are presented in Section 4. Section 5 provides a discussion on our empirical findings. The conclusion, as well as the implications and limitations of the study, are provided in Section 6.

2. Literature Review and Research Hypotheses

2.1. Theoretical Underpinning

The current study draws on the stakeholder theory and social identity theory to formulate hypotheses. Since the 1980s, the stakeholder theory, advocated by Freeman [24], has been widely used by scholars and practitioners to understand CSR [25,26]. Freeman [24] stated that corporate managers must understand the needs of all groups who have a stake in the business (i.e., stakeholders). The essence of any business primarily lies in building relationships and creating value for all its stakeholders, since stakeholders reward or punish corporate behaviour [27]. Companies may suffer both monetary and reputational losses from failing to align their interests with those of their stakeholders [28].

Though the composition of stakeholders may differ depending on company’s industry and business model, the main stakeholders typically include employees, customers, communities, suppliers, and financiers (owners, investors). Under the current research background, the research object of CSR activity response is the consumers. Morsing [29] noted the rise of critical consumer movements demanding that transparency had been an influential factor in the demand and action of CSR. As important stakeholders, consumers need the company to carry out CSR activities and take social responsibility [2], implying that consumers’ satisfaction with CSR affects the demand for products/services [10,30]. However, research has not always supported a positive relationship between companies’ CSR commitment and consumers’ reactions; some studies suggest a positive reaction of consumers towards businesses’ CSR commitment [31,32], while others give evidence that consumers are indifferent to companies’ CSR [33], or are influenced to a limited extent by CSR [34].

We argue that the mixed findings between a company’s CSR commitment and its consumers’ reactions could partly be attributed to the mismatch between the company’s CSR activities and its consumers’ values or concerns. Companies’ CSR activities can target a diversity of issues, while not every consumer has the same concern on all of those issues. Studies also showed that if different dimensions of a company’s CSR activities significantly influence consumers’ responses, the strength and valence of these impacts may differ [15,35,36]. Thus, we argue that when classifying the types of CSR activities, it is logical and meaningful for companies to consider an alignment match between consumers’ CSR needs and consumer values. However, the impact of such matching between consumer values on different CSR activities still lacks in the existing literature.

To understand the mechanism of how such matching/mismatching between a company’s CSR practices and consumers’ values affects consumers’ reactions, the current study introduces a typology of CSR practices and the concept of CCI. It then draws on the social identity theory proposed by Turner and Tajfel [37] to connect CSR and CCI as the theoretical underpinning before developing the hypotheses.

2.2. A Typology of Corporate Social Responsibility Practices

CSR is generally conceptualised as a multidimensional construct [38]. Several typologies coexist in the literature [39,40,41]. The classical framework proposed by Carroll [17] pointed out that social responsibilities should be tied to social issues or topical areas such as consumerism, environment, discrimination, product safety, occupational safety, and shareholders. Following this line of thinking, we categorise a company’s CSR activities into three target types. The first type of CSR engagement targets local community issues. For example, Merck Ltd, Thailand is a subsidiary of Merck KGaA, Germany. It maintains a long-term locallyfocused CSR programme in different regions [42]. Through the local subsidiary, Merck has contributed 20 million Thai Baht to local communities since the programme’s foundation [42]. Such community-focused CSR communicates caring about the needs of, and providing benefits to, the local community. The second type of CSR engagement targets broader environmental issues by delving into macro and global issues such as environmental protection and climate change [43,44,45]. For example, IKEA reduces waste in its value chain and uses environmentally friendly raw materials (e.g., organic cotton and recyclable materials) as much as possible [46,47,48]. The Air New Zealand CSR project, FlyNeutral, encourages passengers to purchase carbon credits, and the collected funds have been invested in New Zealand’s permanent primary forest to offset carbon emissions [49]. The third type of CSR engagement targets broader humanity issues such as violation of human rights (e.g., child labour or gender inequality) or animal rights (e.g., animal cruelty or conducting animal experiments). For example, Starbucks promises to purchase and sell ethically traded coffee [50]. The Body Shop ensures animal cruelty-free policies [51] and has donated approximately 7.5 per cent of its annual profit to charity over the years, which is higher than the donations of many other beauty brands [52]. Therefore, a company could target their CSR using different engagement strategies, i.e., activities that are local-community-focused, environment-focused, and broader humanity-focused when engaging in CSR initiatives.

2.3. Consumer–Company Identification

In recent years, scholars have used CCI to investigate the mechanism through which CSR affects consumer reactions. CCI stresses a social partnership that brings together a shared identity between an organisation and its customers [53]. Consumers tend to identify with companies to fulfil their self-defined needs.

There are different ways that a company could arouse consumers’ identification through their employees, brands, customer groups, special projects, and engagement. Engaging in CSRs is one such way to create CCI. A socially attractive company establishes its association, image, and values via adhering to ethical values, committing to a network with responsible business partners, and connecting with particular ethical communities, thus creating social resources [54]. When consumers see similarity between a company’s socially responsible identity and themselves, they will identify more strongly with the company and formulate positive emotional bonding [55,56]. The emotional interactions between individuals and the corporation will prompt consumers to establish strong, committed, and meaningful relationships with the corporation and become champions of its products [9]. CCI thus fosters favourable consumer reactions such as purchase intention [10], consumer loyalty [11], and positive electronic word-of-mouth (eWOM) [30,57]. Hence, the previous literature shows that CCI mediates the relationship between a company’s CSR practices and consumer reaction such as purchases. As a result, companies implement CSR engagement strategies to establish and strengthen such identification and relationships with their customers.

2.4. Social Identity Theory to Link CSR and CCI

Social identity theory suggests that the individuals identify themselves with a social group due to belongingness and emotional attachment to that group [37]. Scholars have widely used this theory to explain how consumers develop their identities and belongingness with an organization involved in CSR activities [30]. Ma et al. [30] built a framework showing that a company engaged in various CSR activities to manifest its company identity. Consumers then identify with the company as a socially responsible organization to fulfil their need for self-definition of socially responsible consumers.

However, socially responsible consumers are not homogenous as consumers have different personal values which they placed emphasis on. Verma, Chandra and Kumar [18] suggested three foundational consumer values: egoistic, biospheric, and altruistic. Egoistic values focus on the costs and benefits of choices that influence people’s resources, such as wealth, power, and achievement [58]. Previous studies suggested that consumers with high egoistic values care about themselves and personal gains [58,59]. Biospheric values reflect a bigger concern for the environment without a clear link to human beings [60,61]. Altruistic values have been conceptualised as part of a personal values structure or overall guiding principle that motivates individuals to contribute to the well-being of other human beings or care for society as a whole [58,61,62]. A person with high altruistic values benefits others and does not expect external rewards [61,62].

In this study, we argue that a consumer categorizes him/herself and other salient groups into “us” vs. “them”. Such self-categorization could be based on values shared amongst in-group members, constituting social identity. To maintain a positive social identity, people engage in intergroup comparisons that demonstrate a favourable bias toward their in-group, while displaying discriminatory behaviours toward out-groups [63,64,65,66]. In the scenario of choosing a company to identify with, a socially responsible consumer emphasizing particular values will be more likely to choose the company that engages in a certain type of CSR practices that align with his/her values. This is because such a company, though engaging in CSR aligning with a consumer’s values, connects the consumer to the company’s partners and communities sharing similar values. Hence, identifying with such a company will show the consumer’s attachment with the in-group members and enhance the consumer’s own social identity.

Such a strengthened identity will lead the consumer to evaluate the company positively and improves his/her behavioural intentions [10,11,30]. Based on such theoretical underpinning, we develop the hypotheses in Section 2.5.

2.5. Hypotheses Development

Specifically, in line with the social identity theory, consumers with different values may select, identify, or associate themselves with different CSR practices that companies engage in to construct, support, and even enhance their own identities. Local-community-focused CSR, such as support of local non-governmental organisations (NGOs) or NGO partners in providing local community support, financial assistance, or merely demonstrating a genuine concern for the community where the business operates [32,67], represents the organisation’s initiatives of serving and benefiting the local community. Egoistic values focus on the costs and benefits of choices that impact social power, wealth, authority and influence, and such consumers care about themselves and their personal gains [58]. Perlaviciute, Steg and Sovacool [68] argued that egoistic values focus on self-display concerns of nearby projects in contrast to altruistic and biospheric values that focus on others and the environment. Association with care and benefits for the local community where the product is made may enhance the CCI of egoistic consumers. Consumers with stronger self-enhancement or egoistic values may consider the impacts as achievement of their own actions. Based on the above argument, we propose the following hypothesis:

Hypothesis 1 (H1).

CSR practices targeting the local community will have a stronger positive influence on the CCI of high-egoistic consumers than that of other groups.

Companies can also engage in CSR activities targeting environmental protection. Environment-focused CSR concentrates on the organisation’s roles in environmental protection, ecology, climate change, and responsible industrial waste management [43,44,45]. Biospheric consumers have a vital concern for the environment and environmental issues. When companies are involved in such environment-focused CSR, they communicate their care about the earth and environment, establishing a company identity of being environmentally friendly. Such an identity aligns more closely with the needs of biospheric consumers who wish to enhance their own identity by being loyal to and purchasing products/services from the companies. Hence, we propose hypothesis 2:

Hypothesis 2 (H2).

CSR practices targeting the environment will have a stronger positive influence on the CCI of high-biospheric consumers than that of other groups.

Shao, Mahmood and Han [69] argued that altruistic values are associated with the collective welfare of society. Altruistic consumers tend to care for society as a whole and the well-being of other human beings [58,61,62]. Some CSR engagements do not specifically focus on the community where the company operates nor on environmental benefits per se, but instead on broader humanity issues such as gender equality, prevention of child labour, and ethical sourcing. Broader humanity-focused CSR helps establish a company identity of being altruistic and sacrificing profit for the well-being of humanity as a whole. Altruistic consumers may be more likely to find similarities between these companies and themselves and formulate their CCI accordingly. Hence, we propose the following hypothesis:

Hypothesis 3 (H3).

CSR practices targeting broader humanity issues will have a stronger positive influence on the CCI of high-altruistic consumers than that of other groups.

Wu and Tsai [70] found that a high degree of CCI has a favourable impact on consumer purchase intentions measured by proxies of consumer recruitment and loyalty, tolerance of defects, and consumer advice and complaints. According to de Oliveira Duarte and Silva [71], consumers’ identification with the company’s cause and consequent changes in their attitude can predict their intention to purchase. Once a consumer identifies with certain CSR practices of a company, the resulting emotional interaction will prompt the consumer to put more effort into voluntarily realising the company’s target and become a champion of the company [9]. A considerable number of studies suggest that a positive CCI will contribute to a more favourable view of the company’s brand, leading to a stronger purchase intention among the consumers that identify with the company [32,72,73]. Based on the above literature discussion and the social identity theory, we propose the final hypothesis:

Hypothesis 4 (H4).

A stronger CCI will lead to a higher purchase intention.

Figure 1 presents our theoretical model and the four associated hypotheses.

3. Methodology

3.1. Sampling

An online questionnaire survey was administered to understand how a consumer would identify with a company’s CSR and capture the extent of the consumer’s egoistic, biospheric and altruistic values. Ethics approval (H8399) was obtained from James Cook University, where consumers over 18 years of age were targeted using word-of-mouth convenience sampling and snowball sampling. We distributed the questionnaire link through email contacts and online social networks such as Facebook, LinkedIn, and the University’s student affairs department in Singapore. We chose to use convenience sampling to collect data. A standard definition of convenience sampling is researching those elements of the population that are readily available to the researcher [74]. It is a non-probability sampling method. It allows the researcher to observe general consumer habits, opinions, and viewpoints cost-efficiently [75]. It is suitable when the purpose of the research is to test the theory to increase the understanding of their underlying processes [76], more than a generalization of findings to the whole population.

Given that this study aims to test the hypotheses to better understand the mechanism through which consumers react to companies’ CSR practices, convenience sampling will be appropriate rather than describing the characteristics of a general population. Snowball sampling was subsequently used to increase the sample size by encouraging participants to share the link with their social networks. An influential body of social science is based on convenience sampling [77,78], and it has proven sufficient to test psychological theories across various topics [79,80,81,82]. The population of Singapore is 5,704,000. Based on a confidence level of 95% and a confidence interval of 10, the required sample size was 96. In the end, a total of 217 valid questionnaires were collected.

3.2. Survey Design

The questionnaire was developed in English and pre-tested with a small group of participants to avoid uncertainties and ambiguities. The questions are shown in Appendix A. We asked each respondent to choose one company that had purchased products/services from a given list of eight companies: IKEA, ZARA, H&M, Haidilao, Singapore Airlines, McDonald’s, Huawei and Apple. These companies were chosen for two reasons. First, they are well known to Singaporeans. Second, the eight companies have engaged in at least one type of CSR activity based on their official websites, and Facebook and Instagram pages. In

Table 1. Categorisation scheme for CSR posts.

Category	Example
Local-community-focused CSR	“Dear valued customers, we deeply regret to inform that due to recent virus transmission, Haidilao Singapore has decided to suspend the operation of children playground and manicure service temporarily from 28 January 2020. We will announce the availability when the operation resume. We thank you for your continuous support and sincerely apologise for any inconvenience caused.”
	“While ensuring steady operation, we genuinely appreciate everyone who has supported us and never forgot our original aspiration. We actively fulfil our corporate social responsibilities, encourage our employees to participate in charity and volunteer work, contributing to local communities. https://www.haidilao.com/sg/gyhdl/shzr/index.html” (accessed on 12 February 2021)
Environment-focused CSR	“In support of Earth Hour, our stores will be turning exterior lights off for an hour tonight at 8.30 pm.”
	“1.4 billion re-sealable plastic bags make a difference, helping us to reduce our carbon footprint.”
Broader humanity-focused CSR	“Recently, we have received some enquiries on whether any preventive measures have been taken to protect the health and safety of our customers and employees. Below are the measures that we have been practising: …”
	“Here at McDonald’s, it’s an absolute joy and delight to bring a smile on a customer’s face. Here’s one* of many heartfelt letters we receive from customers that inspires us to continue serving with hospitality from our hearts. Ultimately, making McDonald’s a “Happy Place” for everyone is what makes our day.”

, we provide two examples for each CSR. One author collected the posts and categorised them into three types of CSR. Another author checked the categorisation. When the two authors disagreed on categorising a post, the third author became involved in the discussion to decide the best category.

Table 2. The number of CSR posts.

Company	Frequency	Environment-Focused CSR Post	Local-Community-Focused CSR Post	Broader Humanity-Focused CSR Post
IKEA	49	12	1	4
ZARA	11	2	0	0
H&M	13	18	0	3
HAIDILAO	24	0	5	1
Singapore Airlines	15	1	3	3
McDonald’s	67	1	0	6
HUAWEI	8	4	4	1
Apple	28	2	0	3

summarises the number of posts related to three types of CSR activities that these companies engaged in between September 2020 and April 2021.

We then asked participants to rate the selected company’s CSR engagement (environment-focused, local-community-focused, and broader humanity-focused), company ability, and intention to purchase its products and services. The respondents were assessed on their individual consumer values (egoistic, biospheric and altruistic) and were asked to provide some demographic information. A profile of the samples on a range of background characteristics is provided in

Table 3. Demographic characteristics of the survey sample (n = 217).

Characteristic	Classification	Frequency (Number)	Percentage (%)
Gender	Male	134	61.8
	Female	83	38.2
Age (years)	18–25	41	18.9
	26–30	40	18.4
	31–35	43	19.8
	36–40	31	14.3
	>40	62	28.6
Education	Below bachelor’s degree	40	18.4
	At least bachelor’s degree	177	81.6
Working situation	Full-time employed	119	54.8
	Part-time employed	40	18.4
	Seeking work	9	4.2
	Retired	5	2.3
	Home duties	11	5.1
	Student	23	10.6
	Others	10	4.6
Monthly expenditure (excluding housing cost, SGD)	<400	10	4.6
	401–600	27	12.5
	601–800	28	12.9
	801–1000	23	10.6
	1001–1500	27	12.5
	1501–2000	26	11.9
	2001–3000	33	15.2
	>3000	43	19.8

.

3.3. Measurement

The measurement scales for egoistic, biospheric and altruistic consumer values were adopted from Verma, Chandra and Kumar [18]. Three types of perceived CSR were measured by combining the measurement scales from Fatma, Rahman and Khan [32] and Ehsan et al. [43]. The measurement scale for CCI was derived from Pérez and Rodriguez del Bosque [1]; five of six items were adopted to measure CCI, ensuring sufficient validity. Purchase intention was measured using the scale developed by Deng and Xu [31]. Other than these main variables, we also included control variables. Three items measured company ability by combining the measurement scales from Murray and Vogel [83] and Becker-Olsen, Cudmore and Hill [84]; Marín and Ruiz [85]: ‘I believe this company is financially strong’, ‘I believe this company has good management’, and ‘I believe this company makes quality and innovative products’. All responses in the survey used a five-point Likert scale (1 = strongly disagree to 5 = strongly agree).

3.4. Validity Test of the Measures

SmartPLS reports the convergent validity of the measures, as shown in

Table 4. Measurement validity and reliability.

Variable	Item	Corrected Item-Total Correlation	Cronbach’s Alpha	Composite Reliability	AVE
Egoistic Values	EGO1: Social power; I like to have control over others, dominance is important to me. EGO2: Wealth; material possessions and money is important to me. EGO3: Authority; the right to lead or command is important to me. EGO4: Influential; having an impact on people and events are important to me.	0.691	0.780	0.858	0.604
		0.746
		0.830
		0.479
Biospheric Values	BV1: Preventing pollution; conserving natural resources is important to me. BV2: Respecting the earth; harmony with other species is important to me. BV3: Unity with nature; harmonising with nature is important to me. BV4: Protecting the environment; preserving nature is important to me.	0.780	0.853	0.899	0.689
		0.623
		0.737
		0.913
Altruistic Values	AV1: Equality; equal opportunity for all people is important to me. AV2: The world at peace; freedom from war and conflict is important to me. AV3: Social justice; correcting injustices and caring for the weak is important to me. AV4: Helpful; working for the welfare of others is important to me.	0.771	0.813	0.876	0.640
		0.731
		0.729
		0.650
Local- Community focused CSR	LSR1: This company engages in philanthropy contributing to causes such as the arts, education and social services in Singapore. LSR2: This company plays a role in the local community beyond generating economic benefits in Singapore.	0.823	0.778	0.899	0.817
		0.773
Environment-focused CSR	ECSR1: This company adopts good practices in the use, purchase and production Of environmentally friendly products. ECSR2: This company is concerned with respecting and protecting the natural environment.	0.920	0.921	0.962	0.927
		0.928
Broader Humanity focused CSR	BCSR1: This company supports fair-trade practices globally. BCSR2: This company employs women, minorities and disabled persons globally and adopts good employment practices. BCSR3: This company is concerned with improving the general well-being of global society.	0.769	0.743	0.837	0.638
		0.379
		0.907
CCI	CCI1: I share a similar identitywith this company. CCI2: This company fits my personality. CCI3: I feel good being a customer of this company. CCI4: I like saying that I am a customer of this company. CCI5: I feel closely linked to this company.	0.711	0.881	0.913	0.678
		0.775
		0.847
		0.810
		0.713
Purchase intention	PI1: I consider this company as my first choice to buy products or services. PI2: I will buy most relevant products or services from this company. PI3: I am more willing to try new products and services launched by this company.	0.839	0.849	0.909	0.768
		0.745
		0.836
Company ability	CA1: I believe this company is financially strong. CA2: I believe this company has good management. CA3: I believe this company makes quality and innovative products.	0.735	0.777	0.871	0.694
		0.766
		0.706

. Cronbach’s alpha coefficients are all above 0.7, supporting the scale’s internal consistency reliability. The composite reliability was higher than 0.6 [86], suggesting adequate convergent validity of the construct. We also examined the internal validity of the measurement model through average variance extracted (AVE). All AVE values were within the acceptable threshold (a value above 0.7 is considered good, while the value of 0.5 is acceptable) [86]. Therefore, the convergent validity of the constructs was established.

To further establish the discriminant validity across all the constructs, we examined the cross-loading reported in SmartPLS. All the cross-loadings are less than the loading on the primary construct. We also referred to the Fornell and Larcker [86] and the Heterotrait-monotrait ratio of correlations (HTMT ratio) reported in SmartPLS, since Voorhees et al. [87] demonstrate that the AVE-SV (shared variance) comparison [86] and HTMT ratio [88] with 0.90 cut-offs provide the best assessment of discriminant validity and should be the standard for publication in marketing. As shown in

Table 5. Discriminant analysis: Fornell and Larcker Criterion.

Variable	Altruistic	Biospheric	CCI	Egoistic	ECSR	LCSR	PI	CA	BCSR
Altruistic	0.800
Biospheric	0.219	0.830
CCI	0.205	0.255	0.824
Egoistic	−0.103	0.081	0.318	0.777
ECSR	0.022	0.327	0.485	0.101	0.963
LCSR	0.130	0.103	0.321	0.217	0.328	0.904
PI	0.209	0.189	0.742	0.248	0.317	0.301	0.876
CA	0.274	0.238	0.609	0.222	0.314	0.346	0.643	0.833
BCSR	0.185	0.143	0.450	0.320	0.410	0.529	0.360	0.416	0.799

Note: PI = purchase intention, ECSR = environment-focused CSR, LCSR = local-community-focused CSR, BCSR = broader humanity-focused CSR, CA = company ability.

, the AVE for each factor was larger than the square of the correlation estimates of the factor with all other constructs. In

Table 6. Discriminant validity test: heterotrait-monotrait ratio of correlations (HTMT ratio).

Variable	Altruistic	Biospheric	CCI	Egoistic	ECSR	LCSR	PI	CA	BCSR
Altruistic
Biospheric	0.264
CCI	0.234	0.282
Egoistic	0.157	0.138	0.375
ECSR	0.054	0.359	0.538	0.119
LCSR	0.168	0.116	0.386	0.275	0.379
PI	0.245	0.222	0.853	0.304	0.358	0.368
CA	0.346	0.293	0.728	0.291	0.372	0.451	0.785
BCSR	0.277	0.141	0.475	0.337	0.416	0.667	0.418	0.513

Note: PI = purchase intention, ECSR = environment-focused CSR, LCSR = local-community-focused CSR, BCSR = broader humanity-focused CSR, CA = company ability.

, the HTMT is all less than 0.90. Therefore, the discriminant validity is established. We were confident that the measurement model was satisfactory considering all these indicators. Thus, moving on to conduct structural equation modelling was feasible.

4. Results

4.1. Descriptive Statistics

The variable means for the eight companies are presented in

Table 7. Descriptive statistics.

Company	N	PI	CCI	Perceived Overall CSR	ECSR Perception	LCSR Perception	BCSR Perception	Egoistic	Biospheric	Altruistic
IKEA	49	3.728	3.763	3.628	4.153	3.316	3.500	3.459	4.347	4.102
ZARA	11	3.091	2.873	2.909	3.273	2.864	2.697	3.000	4.159	3.932
H&M	13	3.128	3.415	3.121	3.731	2.731	2.974	3.577	4.192	3.615
HAIDILAO	24	3.917	3.792	3.405	3.104	3.438	3.583	3.792	4.000	4.042
Singapore Airlines	15	4.200	3.773	3.476	3.300	3.533	3.556	3.650	4.317	4.183
McDonald’s	67	3.498	3.242	3.405	3.119	3.403	3.597	3.504	3.978	4.187
HUAWEI	8	3.292	3.675	3.482	2.813	3.438	3.958	3.844	3.938	4.469
Apple	28	3.905	3.786	3.383	3.232	3.571	3.357	3.768	3.929	4.089

Note: PI = purchase intention, ECSR = environment-focused CSR, LCSR = local-community-focused CSR, BCSR = broader humanity-focused CSR.

. The results showed that the customers of IKEA, ZARA, H&M, and Singapore Airlines typically held the highest score for biosphere values, while customers of Haidilao, McDonald’s, Huawei, and Apple reported the highest score for altruism values. Interestingly, among the four companies with the highest consumer score for biospheric values (IKEA, ZARA, H&M, and Singapore Airlines), the environment-focused CSR perception of participants was higher than the perception of local-community-focused CSR and broader humanity-focused CSR, except for Singapore Airlines. A similar situation was evident among the four companies with the highest scores for altruistic values. Only Apple did not show a higher broader humanity-focused CSR perception, while the remaining three companies (Haidilao, McDonald’s, and Huawei) showed a higher value for broader humanity-focused CSR perception than the other two types of CSR perception.

4.2. Hypotheses Testing

We employed SmartPLS for partial least squares structural equation modelling (PLS-SEM) to test our hypotheses. PLS-SEM overcomes model identification problems, and it is a powerful method for analysing complex models using smaller samples [89,90]. The PLS-SEM approach does not require normal data distribution and accommodates small samples size [91]. The approach can have categorical to ratio data, and hence has the advantage over covariance-based SEM, which has a strong assumption of multivariate normality, requires a large sample size, can only include interval or ratio data and is not suitable for complex models [92]. PLS-SEM is a better approach than regression as the former operates much like multiple regression analysis but could handle the latent variables [93]. The model tested in this study was complex because it contained nine main latent variables with thirty-five indicators and the mediation effects (one mediator) and moderating effects (three moderating effects were tested in each model). Based on the sample size calculator for structural equation models (https://www.danielsoper.com/statcalc/calculator.aspx?id=89, accessed on 15 February 2021), 213 observations were the required minimum sample size. Our sample size was 217, just surpassing the threshold. As a result, PLS-SEM was appropriately used in our investigation. PLS-SEM results are shown in

Table 8. PLS-SEM results.

Variables	Model 1		Model 2		Model 3		Model 4
	1a: CCI	1b: PI	2a: CCI	2b: PI	3a: CCI	3b: PI	4a: CCI	4b: PI
Age	0.125 * (0.059)	0.097 (0.060)	0.117 + (0.060)	0.036 (0.063)	0.100 + (0.058)	0.030 (0.054)	0.115 + (0.064)	0.026 (0.053)
Egoistic	0.200 ** (0.061)	0.121 (0.070)	0.165 ** (0.058)	−0.006 (0.057)	0.181 * (0.057)	0.012 (0.054)	0.218 ** (0.060)	0.017 (0.054)
Biospheric	0.002 (0.063)	−0.003 (0.065)	0.005 (0.064)	−0.027 (0.061)	0.056 (0.059)	−0.003 (0.063)	0.007 (0.063)	0.004 (0.059)
Altruistic	0.075 (0.057)	0.038 (0.062)	0.081 (0.057)	0.007 (0.050)	0.081 (0.057)	−0.008 (0.053)	0.061 (0.059)	−0.010 (0.048)
LCSR	−0.020 (0.069)	0.020 (0.068)	−0.080 (0.069)	0.033 (0.055)	0.028 (0.067)	0.030 (0.064)	−0.022 (0.069)	0.033 (0.063)
ECSR	0.326 *** (0.066)	0.130 * (0.065)	0.332 *** (0.060)	−0.047 (0.063)	0.244 *** (0.064)	−0.060 (0.065)	0.322 *** (0.063)	−0.061 (0.067)
BCSR	0.081 (0.073)	0.031 (0.077)	0.132 (0.079)	−0.008 (0.062)	0.118 (0.075)	−0.015 (0.062)	0.109 (0.072)	−0.017 (0.068)
Company ability	0.418 *** (0.068)	0.552 *** (0.060)	0.427 *** (0.069)	0.354 *** (0.061)	0.384 *** (0.066)	0.289 *** (0.055)	0.419 *** (0.065)	0.295 *** (0.057)
Egoistic * LCSR (H1)			0.151 * (0.045)
Biospheric * LCSR (H1)			−0.088+ (0.054)
Altruistic * LCSR (H1)			−0.054(0.062)
Egoistic * ECSR (H2)					−0.089(0.061)
Biospheric * ECSR (H2)					0.162 ** (0.057)
Altruistic * ECSR (H2)					−0.110 + (0.052)
Egoistic * BCSR (H3)							0.094 + (0.054)
Biospheric * BCSR (H3)							−0.037 (0.051)
Altruistic * BCSR (H3)							−0.035 (0.063)
CCI (H4)		0.589 *** (0.061)		0.532 *** (0.060)		0.591 *** (0.064)		0.582 *** (0.063)
R-squared	0.518	0.613	0.559	0.615	0.554	0.614	0.533	0.614
SRMR for saturated model	0.065		0.066		0.065		0.065

Note: *** p < 0.001; ** p < 0.01; * p < 0.05; + p < 0.1; LCSR = local-community-focused CSR, ECSR = environment-focused CSR, BCSR = broader humanity-focused CSR, PI = purchase intention.

.

In light of our conceptual framework, four sets of PLS-SEM analyses were run to test the hypotheses. In Model 1 (1a and 1b), we only included the main effects of key variables and control variables, age and company ability, to test how CCI mediated the relationship between the main variables (types of CSRs and types of consumer values) and purchase intention. In Model 2 (2a and 2b), we included the interactions between local-community-focused CSR and the three consumer values to test the moderation effects of consumer values on the relationship between local-community-focused CSR and CCI. In Model 3 (3a and 3b), we included the interactions between environment-focused CSR and the three consumer values to test the moderation effects of consumer values on the relationship between environment-focused CSR and CCI. Model 4 (4a and 4b) included the interactions between broader humanity-focused CSR and the three consumer values to test whether some consumer values strengthened or weakened the relationship between broader humanity-focused CSR and CCI.

In Model 2a, the interaction between egoistic values and local-community-focused CSR was significantly positive (β = 0.151; p < 0.05). In addition, the interaction between biospheric values and local-community-focused CSR was significantly negative (β = −0.088; p < 0.1), while the interaction between altruistic values and local-community-focused CSR was insignificant. These findings support H1, suggesting that local-community-focused CSR has a stronger positive influence on CCI for consumers with higher egoistic values than that of the other two groups.

In Model 3a, the interaction between biospheric values and environment-focused CSR was significantly positive (β = 0.162; p < 0.01), while the interaction between altruistic values and environment-focused CSR was significantly negative (β = −0.110; p < 0.1), and the interaction between egoistic values and environment-focused CSR was insignificant. These results support H2, suggesting that environment-focused CSR has a stronger positive influence on CCI for consumers with higher biospheric values than for the other two groups.

H3 predicts that a broader humanity-focused CSR has a stronger positive influence on CCI for consumers with higher altruistic values than for the other two groups. In Model 4a, the interaction between egoistic values and broader humanity-focused CSR was significant and positive (β = 0.094; p < 0.1), while the interaction between biospheric values and broader humanity-focused CSR and that between altruistic values and broader humanity-focused CSR were not significant. These results do not lend support to H3. Instead, our results imply that holders of egoistic values identify more with companies committed to broader humanity-focused CSR.

H4 predicts that a stronger CCI leads to a higher purchase intention. According to our PLS-SEM analysis, CCI does have a significantly positive effect on purchase intention in all four models, 1b, 2b, 3b and 4b (p < 0.001), thus supporting H4.

In addition to the PLS-SEM test, we also performed linear regression using SPSS to test our hypotheses. The regression results were highly consistent with the findings from PLS-SEM, which further confirmed the credibility of our results.

5. Discussion

Noting the mixed findings between a company’s CSR commitment and its consumers’ reactions, this study draws on the stakeholder theory and social identity theory to examine how consumers with different values formulate CCI with a company that engages in various CSR activities. Our results echo those of previous studies and, at the same time, provide new insights into the CSR literature.

Some research found a positive effect of CSR engagement on CCI [11,12,13,31]. This effect has been tested by examining the general CSR statement about a company. For example, in Huang, Cheng and Chen [11] or Deng and Xu [31], a company’s CSR engagement was measured by respondents’ perception (e.g., this company is socially responsible), without differentiating actual CSR types or targets. Although Lii and Lee [13] differentiated three types of CSR, they only investigated the execution style of CSR in the forms of philanthropy, cause-related marketing, and sponsorship. However, the classical framework proposed by Carroll [17] pointed out that social responsibilities should be tied to social issues or topical areas such as consumerism, environment, discrimination, product safety, occupational safety, and shareholders. Based on the stakeholder theory, failure to address important issues that consumers care about will disappoint and drive them away [2]. As a result, topical issues which a company should focus on become an important strategic decision to direct precious resources into CSR. This study provides solutions to such a decision by categorising companies’ CSR engagements into three types based on the topical targets, i.e., local-community-focused, environment-focused, and broader humanity-focused CSRs. Our results suggest that not all of them lead to a positive response from their consumers.

Specifically, we find that environment-focused CSRs generally enhance CCI for consumers, and such a positive relationship between environment-focused CSRs and CCI is notably stronger for those consumers holding strong biosphere values. Such a result is generally consistent with prior studies, which show a positive relationship between environment-focused CSRs and consumer reaction. For example, Louis, Lombart and Durif [94] gave evidence that retailers’ environment-focused CSRs positively impact consumers’ loyalty. Hanson et al. [95] revealed that brands that sell goods receive positive responses from consumers when prioritising environmental activities over social activities. Such a finding may be attributed to companies typically marketing their environment-related CSRs more obviously and continuously, and pro-environment behaviour becoming a modern and fashionable identity with which many consumers hope to associate themselves.

Concerning companies’ engagement in local-community focused CSR, Li, Zhang and Sun [96] found that Chinese consumers rated low importance on companies’ CSR engagement in sponsoring the local community, caring for the vulnerable, and public social activities. Our results, to some extent, echo their findings by showing that not all consumers identify with a company engaging in local-community-focused CSR; instead, local-community-focused CSR seems to be effective only for egoistic consumers in forming identification (i.e., CCI). This is understandable because egoistic consumers relate their personal benefits better with companies committed to their local community, and hence more responsive to such a type of CSR.

When engaging with broader humanity-focused CSR, we find that these activities significantly influence the identification (i.e., CCI) of egoistic consumers but not of altruistic consumers. Egoistic values are similar to self-enhancement and focus on impacts for the individual alone. Self-enhancement is a concern for oneself and includes the values of power, achievement, and hedonism [97,98]. Hence, consumers with high egoistic values may like to display their power to others and make an impact on other people’s lives. Broader humanity-focused CSR is concerned with improving the general well-being of global society, and it is likely to be linked to an ability to exert influence on people.

Finally, our results show that a stronger CCI leads to a higher purchase intention for the company’s products and services. Combining with the results of the interaction between the types of CSR activities and the types of consumer values, we show a conditional mediating effect of CCI in explaining the impact of CSR on consumer reactions. Only when a company’s CSRs align with its consumer’s values can an identification be formulated, which leads to the intent to purchase the goods and/or services. Otherwise, a company’s engagements in CSR will not be able to drive CCI, let alone purchase intention. This is different from the previous studies, which show a general mediating role of CCI. For example, Fatma et al. [57] revealed that banks’ level of CSR engagement positively influences CCI on social media, which leads to positive eWOM among consumers. Ma et al. [30] gave evidence that CCI mediates a company’s CSRs and eWOM. Huang, Cheng and Chen [11] found that companies’ CSR affects CCI, increasing customers’ brand loyalty. Considering that those studies measured CSR using a general statement, instead of differentiating targeting issues of CSRs, our results could be more refined.

In summary, our results show a mechanism through which a company engaging in different types of CSRs, creates CCI and higher purchase intention for consumers with different types of values. Such results, thus, highlight the importance of aligning the targeted CSR issue with the selected consumer groups.

6. Conclusions, Implications and Limitations

Noting that little, if any, research has previously been conducted to reveal the process of forming a match between a company’s CSR engagement and a consumer’s values identity during CCI formulation, we used eight well-known global and local brands to examine how different types of CSR activities influence CCI in a tropical multicultural setting. By drawing on the stakeholder theory and social identity theory, our results show that consumers do not respond positively to all types of CSR activities. Instead, their responses to a company’s CSR are dictated by their own values. Specifically, consumers with egoistic values are more likely to formulate a CCI with a company engaging in local-community-focused CSR and broader humanity-focused CSR, such as fair-trade initiatives, while consumers with biospheric values tend to formulate a CCI with a company engaging in environment-focused CSR. We also find that consumers formulating a stronger CCI with a company demonstrate a higher purchase intention towards the company’s products/services. In contrast, consumers may not identify with a CCI when the company’s CSR does not align with consumers’ values. That is, CSR carried out by companies may be meaningless to consumers. Such results confirm that it is imperative for a company to choose and select the suitable types of CSR activities to address the target consumers’ values, so as to satisfy such stakeholders’ needs and positively influence their purchase behaviour. Thus, these findings bring new insights into the literature on both theoretically and practically consumer responses to CSR.

6.1. Theoretical Implications

Our results enrich the CSR literature by showing how a company’s CSR influences consumer reaction. Prior research reports mixed findings between a company’s commitment to CSR and its consumers’ responses [31,33]. Some studies applied the CCI concept to explain the mechanism through which CSR influenced CCI and consumers’ behaviour and revealed that the effectiveness of such mechanism is contingent on companies’ choice of CSRs [10]. We extended this line of research by looking into the two contingencies, i.e., heterogeneity of companies’ CSR engagement and consumers’ values. We introduce a typology of CSR engagements based on Carrol [17] and categorize CSR activities into three types: local-community-focused CSR, environment-focused CSR and broader humanity-focused CSR. We also categorize consumer values into three types: egoistic, biospheric, and altruistic, following Verma, Chandra and Kumar [18]. We then drew on the social identity theory and demonstrated a mediation effect moderated by consumer values: the consumers with different values identify with a company engaging in some specific type of CSR. They formulate CCI and enhance their identity and hence have a higher purchase intention.

The present research, thus, on the one hand, enriches the stakeholder point of view in understanding CSR by showing that socially responsible consumers, an influential group of stakeholders, are not homogeneous but differ in terms of what they value and seek. Companies hence have to align their CSR activities with the socially responsible consumers’ values and concerns. On the other hand, the mechanism explored in this study also refines the application of social identity theory in CCI research. Ma et al. [30] applied the social identity theory and showed that a company engaged in various CSR activities to manifest its company identity; consumers then identify with the company as a socially responsible organization to fulfil their need for self-definition of socially responsible consumers. Our results show that there could be sub-groups of socially responsible consumers. To serve as a bridge of bonding like-minded socially responsible consumers, a company has to choose the type of CSR to engage in carefully.

6.2. Practical Implications

Practically, this study provides practitioners and managers with insights into developing the best CSR practices in the tropics by using Singapore’s experience as an example. Starting from the typology of CSR activities, the main results show that companies could engage in three main types of CSR activities, as shown by the eight selected companies in this study; however, not all of them could address their important stakeholders, consumers’ concerns equally. Our study suggests that the target customers of different companies have different value preferences. For example, in our sample, the customers of IKEA, ZARA, H&M, and Singapore Airlines show high biosphere values, while customers of Haidilao, McDonald’s, Huawei, and Apple demonstrate high altruism values. Nevertheless, not all these companies have aligned their CSR engagements with their target consumers’ values. For example, though Singapore Airlines’ customers indicate high biosphere values, the company has significantly fewer activities (proxied by postings on their social media channels) on environment-focused CSR rather than the other two types of CSRs. This could make their CSR communication less effective in building CCI with consumers and driving consumer behaviours. Considering limited resources that companies could devote to CSRs, both local and global corporations must be able to understand the value orientations of their target customers and adopt appropriate CSR to attract them and gradually cultivate the CCI of the diverse customer base.

Such implications are particularly important in formulating CSR strategies in tropical regions. Tropical ecosystems are at the nexus of biodiversity loss, climate change, and emerging and persistent diseases and, hence, consumers tend to be more sensitive to the environmental issues [99]. This is, in fact, shown in our selected companies, in that consumers of all the selected companies have indicated a high score for biosphere value. As a result, companies in tropical regions are recommended to engage in and show more cases related to environment-focused CSR. For example, we recommend that companies engage in and communicate more about their environment-focused CSR, e.g., sustainable production and logistics process, reusing and recycling resources or reducing resource expenses, which are critically important concerns in tropical regions. At the same time, many tropical countries are less industrialized and less developed [100], whereby local consumers in under-developed economies may hold higher egoistic values, i.e., value their personal resources. Companies could thus engage more with the local communities by providing pro bono services to the neighbourhood, organizing activities to interact with the local people, and inviting the local community as ambassadors. Together with other posts related to environmental protection and local community well-being, such engagements could be regularly posted on their social media channels (official websites, Facebook, YouTube, and Instagram) or conveyed in other marketing campaigns to enhance the CCI of the two groups (biosphere and egoistic) consumers.

6.3. Limitations and Future Research Direction

Limitations of this study include, first, possible social desirability bias arising from self-reporting. Social desirability bias is defined as ‘a need for social approval and acceptance and the belief that this can be attained by means of culturally acceptable and appropriate behaviours’ [101]. Although we used an anonymous online survey to minimise social desirability bias [102,103,104,105], previous studies have reported that CSR-related consumer research is more likely to suffer from this form of bias [106], in that participants may tend to exaggerate their interest in CSR during surveys because CSR is the ‘right’ thing to do [107,108]. Such social desirability bias could also explain our result of the majority of respondents scoring lower for egoistic values than the other two types of values. Second, the list of companies we provided for participants to choose from in the survey included those operating in multinational contexts and which almost all consumers know well because we hoped to capture the actual customers of the companies. Therefore, caution is necessary when extending the results to small- and medium-sized enterprises. Third, we used convenience sampling to collect consumers’ perceptions. The convenience sample included some of the researchers’ personal contacts, the local community, and the online community (e.g., some forums and social media platforms such as LinkedIn, Facebook, and Instagram). As a result, some groups of consumers may be over-represented. More specifically, in 2021, 61.9% of the population of Singapore was aged between 20 and 64 years [109], while 71.4% of the respondents in our sample were 18–40 years old. In addition, 61.8% of the respondents were male, higher than the ratio for Singapore’s whole population (50.42%). Lastly, our sample is from Singapore, the only developed country in South-East Asia. Therefore, our results may not be able to represent the actual situation of developing countries in the region. Future research should test neighbouring developing countries and compare the results.