The Impact of Communicating Sustainability and Ethical Behaviour of the Cosmetic Producers: Evidence from Thailand

Abstract

Purpose—This study aims to demonstrate the antecedent factors of consumers’ cosmetics purchasing behaviour in the emerging market of Thailand from the perspective of sustainability. Specifically, the study aims to quantitatively analyse the impact of three hypothesised antecedents of consumer behaviour: product quality, communicating sustainability, and ethical business behaviour. Methodology—A quantitative methodology is applied in the study, which collects survey data from Thailand. This study focuses on two cosmetic brands in Thailand, a domestic brand and an international brand. The total 800-sample dataset was analysed using Structural Equation Modelling to validate a conceptual model with measurements of three antecedent factors: quality, ethical behaviour, and communication sustainability. Findings—It is found that ‘ethical behaviour of the producers had a non-significant impact for all samples and the ‘domestic brand’, whereas communicating sustainability had a significant impact in all sample cases. The proposed measurement scales present a practical and pioneering tool for assessing consumer responses and behaviour towards cosmetic brands. The set of scales will also help cosmetics marketers to appraise their strategic planning and monitor their progress toward creating and identifying consumer loyalty to cosmetics brands via producers’ ethical behaviour and CSR messaging. Originality—The global market and inter-stakeholder communications have greatly changed the way people perceive, behave towards, and react to business suppliers. The understanding of consumer brand loyalty in the cosmetics industry and the business strategies focused on the impact of communicating sustainability with ethical behaviour remain limited, especially in the context of the emerging market. This research contributes to filling this gap with empirical analyses.

1. Introduction

In an era marked by a concern for sustainable development, a closer investigation into the relationship between businesses that aim to sustain their operations through ethical behaviour and the consumers’ responses to this behaviour is crucial in view of the desire to propose more actionable implications for both businesses and consumers. Many authors have noted that decision-making today is more complex than at any time in the past, especially given the vast array of different products with which consumers are confronted [1,2]. Meanwhile, the notion of corporate social responsibility (CSR) has become one of the most popular topics of discussion among academics and marketers alike, specifically in relation to its impact on consumer decision-making [3,4].

It is also important from a CSR perspective to disclose manufacturing processes and appropriate ingredient lists as evidence that a company is engaged in ethical business practices, as well as to ensure that consumers can use products safely and without risk to their health. While various academic discussions have outlined the clear implications here, these discussions have largely revolved around the international brands and consumer responses in the Western context [4,5]; by contrast, little attention has been paid to consumer responses to the ethical attributes of cosmetic products and business behaviour, especially in emerging markets.

It has been stipulated that a difference exists between the behaviour of consumers in emerging markets and those in mature markets in relation to CSR. In this regard, it is helpful to reflect on research suggesting that consumers’ perceptions of CSR and their interest in ethical corporate behaviour change during the market maturation process [6,7].

Although there is a growing body of relevant research on emerging markets due to their growing populations and potential for expansion, further in-depth research is required to assess the feasibility of data collection and the priorities of companies. The results of this valuable empirical study based on data from an emerging market suggest that, in contrast to the Western view that CSR factors tend to have a strong influence on consumer purchasing behaviour, consumers in emerging markets do tend to be interested in and favour corporate CSR, but this interest tends not to lead to specific behavioural changes [6].

Given the potential size of the market in question, it is vital that companies develop strategies based on in-depth research into consumer behaviour in emerging markets. In particular, the impact of the ethical attributes of cosmetics on consumer behaviour is an important and urgent research topic for companies seeking to attract consumers in emerging markets and develop them as loyal customers. However, there is a lack of research on this subject in academia today, which has led to a lack of basic materials available for use in developing concrete marketing strategies.

In this study, we aim to fill this research gap, to investigate the antecedents that determine cosmetic choice behaviour in emerging countries from an ethical perspective, and to elucidate the associated consumer behaviour patterns. For this purpose, we select two different cosmetic brands—a ‘domestic brand’ and an ‘international brand’—and decipher the different phases of consumers’ evaluation of each brand along with their purchasing behaviour. By investigating the antecedent factors that may influence consumers’ loyalty to a particular cosmetic product and developing an analytical framework characterised by practical measurement scales, we aim to contribute to future research on product choice behaviour in emerging economies by providing models and scales for analysing consumer behaviour in these contexts.

2. Literature Review

2.1. Focus of the Study

2.1.1. Context: Thailand as an Emerging Market

In order to make our research more informative, it is important to keep in mind the themes of the Thai cosmetics industry and consumer sentiment. For example, customers in Thailand, a representative emerging market, are known to exhibit tendencies and behaviours distinct from those in the same class of middle-developed countries. In this regard, for example, [7] found that while conspicuous value generally influences consumer willingness to pay for luxury goods, an analysis of survey data collected from Thai consumers showed that conspicuous value does not have a significant effect on purchasing behaviour. In an empirical study based on consumer data from Turkey, also an emerging market, it was found that consumers favourably perceive CSR actions taken by companies, but that this does not necessarily lead to specific purchasing behaviour [6].

Thus, although the impact of CSR and other ethical corporate strategies on consumer behaviour is generally agreed upon and confirmed in the international context, further research into emerging markets is required. It is understood that Thai consumer behaviour is impacted by the attachment to, pride in and respect for one’s own culture and traditions embedded in the mindset of Thai people [8,9,10]. Some empirical research has been conducted based around datasets obtained from a variety of emerging markets [11,12,13]. However, analytical conclusions on this subject tend to vary such as those related to, e.g., the impact of CSR behaviour. In the business field—including the specific area of cosmetic purchasing behaviour, the subject of this study—empirical research into CSR among consumers in emerging countries is particularly lacking with regard to comparisons between home and international brands. The issues addressed by this study may accordingly contribute to stimulating further debate in the relevant fields.

2.1.2. Research Target: Cosmetics in Thailand

Thailand has a distinctive demographic composition among emerging countries, with a relatively high proportion of Gen-Y and Gen-Z. This means that the population aged between 10–20 years old is currently driving economic activity and consumption in Thailand, a trend that is expected to continue for some time to come [14].

Thailand is also known to be a highly educated country, due to a policy of limiting childbirths introduced in 1970 and the increasing participation of women in society, which has led to an increased preference for higher education to ensure the future of a smaller number of children [15]. The educational level of Thai workers has also shown distinctive trends. For example, in the four years from 2012 to 2016, the proportion of workers with a primary education or lower decreased by 5.6%, while the proportion with a university degree or higher increased by 3.8% across Thailand [14]. The trend towards higher education is particularly noticeable in urban areas, where the proportion of workers with a university degree or higher is much greater: according to Thailand’s most recent available data, as of 2019, the proportion of the labour force that is highly educated (as a percentage of the total labour force) is indicated at 82.8% [14].

Therefore, this study focused specifically on Thai consumers belonging to younger generations with a higher educational background as a leading consumer segment in commodity consumption, aiming to examine and understand the latest market trends in cosmetics consumption, which is a key theme of the study.

Previous research has shown that young consumers are more interested in the ethical aspects of business and are more likely to focus on a company’s CSR strategy and the ethical value of its manufacturing processes when making a purchase. However, related research in emerging markets remains insufficient. It would therefore be useful to quickly conduct research into the behaviour of younger and more educated consumers, who are considered to be particularly sensitive to CSR, in order to improve business ethics around sustainable consumer behaviour and the responsible production of everyday products. This is also important from the perspective of improving business ethics around sustainable consumer behaviour and the responsible production of everyday products.

2.2. Quality: Ethical Sources and Consumer Experiences

Manufacturing technology is one of the key issues for cosmetics producers. Retailers and manufacturers have partnered with supply chain technology organisations to support their visibility and success throughout the cosmetics supply chain [16]. The cosmetics sector has attempted to implement the technology required to secure ethical sources and ensure that ingredient safety standards are met [17]. Ref. [18] developed and presented a framework for cosmetics consumers’ satisfaction with cosmetics brands, including the products’ quality. Moreover, young adult consumers are regarded as the most significant consumer segment in relation to ethical and sustainable consumption [19], which suggests that cosmetics producers and retailers should focus more on the ethical attributes of their products and the ethical behaviour of cosmetics suppliers [20]. As an advanced technology, blockchain technology has also helped to improve supply chain visibility and support ethical sourcing [21].

Having recognised the change in markets driven by technological innovation, cosmetics manufacturers have subsequently formed partnerships with ICT specialists to improve the consumer experience and enhance their operations [22]. A crucial element here is that of analysing product popularity-related data that are based on ingredients and sources. By adapting to this change in the market, a business can enhance its productivity and longevity through the implementation of a digital retailer strategy, use of meaningfully ethical ingredients, and the adoption of fair processing procedures throughout the supply chain network before the products are delivered to the end consumers [1].

The 1223/2009/EK Decree of the European Union requires cosmetics producers to collate and evaluate reports on the disadvantages of the cosmetic products they sell [23]. The EU regulatory framework also emphasises the importance of highly problematic non-desired effects, with guidelines encouraging European cosmetics manufacturers to develop effective strategies to prevent or mitigate these effects in order to remain compliant with the legal requirements for operating a cosmetics business. However, no such scheme exists in the East Asian region; the relevant discussions and experiences in Europe have shed some light on these sectors. For the analytical model, three potential variables are developed regarding the quality of the products, ethical sourcing and consumers’ brand loyalty [24,25,26,27,28,29,30,31]. Based on the above discussion, the first hypothesis is as follows:

Hypothesis 1 (H1).

Quality, ethical sourcing, and production processes have an impact on consumer loyalty towards cosmetic products.

2.3. Communicating Sustainability

Brand loyalty reflects the likelihood that a customer will make a repeat purchase or decide to switch to another brand, especially when the original brand undergoes a change [32,33]. Ref. [34] emphasised that businesses need to improve their ability to attract consumers through what they termed ‘the new brand spirit’.

Advertising is a commonly used method of attracting consumers and ensuring their loyalty. Ref. [35] investigated the impact of ethical advertising on consumer loyalty and found that its positive impact results in repeat purchases. It can therefore be argued that brand attachment and consumers’ positive attitudes towards a brand are largely related to the perceived quality of its products [36]. However, focusing on the cosmetics sector, [37] used empirical analysis to determine that only corporate authenticity has a positive effect on brand attachment and consumer loyalty; accordingly, the authors suggested that marketing strategies should be designed based on CSR activities that effectively demonstrate corporate authenticity. Elsewhere, following an analysis of the corporate messages on companies’ websites, [38] asserted that the communication of sustainable business behaviour is a key factor behind consumer loyalty.

The concept of consumer engagement and consumer behaviour in brand communities is a popular research topic in the field [39,40]. Moreover, in the more recently established academic arena of social responsibility and consumers’ responses, the relationships between ‘communicating sustainability’ and ‘brand loyalty’ or consumer engagement have been explored in a variety of contexts and markets. It is always critical to examine the latest patterns of consumer behaviour in the most recent situation and conditions [41,42]; however, most existing research has focused on consumer engagement within brand communities on social media platforms.

Therefore, further discussions in the broader context of market perspectives could be enhanced by expanding the research focus to include arenas outside of social media platforms. Based on the discussions above, the second hypothesis is as follows:

Hypothesis 2 (H2).

Communicating sustainability has an impact on consumers’ loyalty towards cosmetic products.

2.4. Ethical Behaviour of Producers

Thailand is an example of an emerging market in which both the economy and consumer behaviour have been undergoing transitions, with a variety of goods and services on offer that bear an array of brand names and are imported from numerous countries [43]. Due to the huge increase in available consumer choices and the changes in consumer needs and desires, competition in the cosmetics market has become fiercely intense. In light of this, it is crucial that firms understand the factors affecting consumers’ purchasing behaviour [44]. Specifically, it is important to identify distinct triggers that will enhance ethical consumption, along with the response to companies’ ethical behaviour within the marketing context. Ref. [45] empirically investigated consumers’ reactions to animal-tested cosmetic products to evaluate the impact of ethical manufacturer behaviour.

In terms of the cosmetics industry, the behaviours of Asian consumers tend to be vastly different from those in the West; this is especially true in Thailand, due to the numerous differences in culture, physicality, and lifestyle [46,47]. Here, the consumer behaviour model devised by [48] helps to explain how a large number of factors can influence consumption and which relevant behaviours are implicated. The discussion suggests that higher (or lower) levels of ethical behaviour by cosmetics providers can be associated with higher (or lower) loyalty.

Regarding the analytical model, three variables were developed in the context of brands’ ethical behaviour and consumers’ loyalty based on the relevant academic discussions [49,50,51,52,53,54,55].

Based on the above discussion, the third hypothesis is as follows:

Hypothesis 3 (H3).

The ethical behaviour of producers has an impact on consumers’ brand loyalty towards cosmetic products.

2.5. Consumer Decisions and Brand Loyalty

Ref. [56] defined motivation as a basic concept of human behaviour that drives an individual to take a particular action; in other words, motivation refers to the processes that initiate, guide and maintain goal-oriented behaviour [57]. Ref. [58] suggested that potential attendant factors and consequences of specific behaviours are not definitively determined and are both extremely complicated and interrelated. This conclusion has been reiterated by an extensive body of academic material, for example, by [59] in their study based on different datasets from various markets. In another academic source, [60] discussed consumer behaviour towards eco-friendly cosmetic products.

The interactive relationships among the factors underpinning consumers’ satisfaction, decision making, and brand loyalty have been discussed and examined by numerous scholars [61,62,63]. The consumer’s final decision generally involves various stages and elements driven by specific influential factors. In short, consumer behaviour does not represent a linear process. In line with this discussion, [64] also emphasised the benefits of quality management and effective communication with customers to attract more consumers in the markets. Regarding the analytical model, three variables were developed based on academic sources representing brand loyalty measurements [65,66,67,68,69,70,71].

Based on the above discussion, the fourth hypothesis on the theme of co-variant relationships for three antecedent factors leading to brand loyalty emerges as follows:

Hypothesis 4 (H4).

The three main influential factors—quality, communicating sustainability, and ethical behaviour—have interactive relations.

2.6. Conceptual Model with Hypothesised Relationships

In summary, the theoretical arguments underpinning each latent factor and their relationships are listed below in

Table 1. Hypotheses and Supporting Discussions.

Observed Variables	Supporting Academic Sources for Observed Variables		Hypotheses	Supporting Academic Discussions
QTY1: High quality of the products are important	Chinomona et al. (2017); Hwang et al. (2021)	H1	Quality, ethical sourcing and production processes have an impact on consumer loyalty towards cosmetic products.	Amberg and Fogarassy (2019); Arce-Calero et al. (2019); Lores et al. (2016); Marks & Spencer (2020); Montagnini et al. (2016); Prak (2020); Renner et al. (2017); Sorum (2020); Supply Chain Movement (2018)
QTY2: I prioritise the quality of ingredients and ethical sourcing, even the price is higher than others, when purchasing cosmetics	Kassim et al. (2021); Pearce (2018); Vorster (2018)
QTY3: I choose the brand which provides good quality of the products	Chaovalit (2014); Spiteri Cornish (2013); Taghipour et al. (2017)
CS1: Attachment to the brand which has sustainability communication	Gupta et al. (2021); Hwang and Kandampully (2012); So et al. (2013)	H2	Communicating sustainability has an impact on consumers’ loyalty towards cosmetic products.	Boateng et al. (2020); Chung and Sung (2019); Conrad and Thompson (2016); Gómez-Suárez (2019); Jung et al. (2020); Kong et al. (2021); Ranjan and Sahu (2014); Shanahan et al. (2019); Siano et al. (2016); Wirts et al.; (2013); Zhou et al. (2012)
CS2: Feel resonance to the brand with sustainable communication	Abraham and Joseph (2020); Hashem et al. (2020)
CS3: Communicating sustainability is helpful to nurture my loyalty	Park and Kim (2019); Schultz and Block (2015); Zehir (2011)
EB1: I pay attention to the brands’ societal activities	Ahn et al. (2020); Aydın (2019)	H3	The ethical behaviour of producers has an impact on consumers’ brand loyalty towards cosmetic products.	Blackwell et al. (2006); Grappe et al. (2021); Jakubanecs et al. (2019); Sthienrapapayut et al. (2018); Wirunphan and Ussahawanitchakit (2016); Wongprawmas and Canavari (2017)
EB2: I choose brands which behave and act ethically	Ozdemir et al. (2020); Park et al. (2017)
EB3: I choose brands which look after the earth	Abdullah et al. (2019); Dahlmann (2020); O’Faircheallaigh and Ali (2017)
Hypotheses on the covariance relationship between the above three latent factors		H4	Three factors, quality, communicating sustainability, and ethical behaviour, have interactive relations.	Akbari et al. (2016); Asgari and Hosseini (2015); Briliana and Mursito (2017); Gilal et al. (2020); Franzoi, (2011); Liu et al. (2017); Priyo et al. (2019); Sadiq et al. (2021); Zgirskas et al. (2021)
BL1: I will purchase the same brand product again	Chan (2016); Shalehah et al. (2019); Taylor (2004)
BL2: I will purchase the same brand products even if other brands are being sold at discounted price.	Chan and Mansori (2016); Villarejo-Ramos and Sanchez-Franco (2005); Vinh and Phuong (2020)
BL3: I will recommend the brand I like to others	Del Rio et al. (2001); Rizvi and Dahri (2020)

.

Based on the literature review in developing measurements for hypotheses, a model was designed that includes the three main influential factors, namely quality, communicating sustainability, and the ethical behaviour of producers, to be tested and validated using primary data-based analysis. As the figure shows, these three factors are regarded as interrelated, with the attendant relationships converging to result in brand loyalty.

H1 focuses on the quality of the sourced ingredients (where quality is assessed in part based on the transparency and accountability of the sourcing process). Quality should be one of the key factors leading to consumer loyalty, as it is one of the established antecedents of brand loyalty among purchasers of cosmetics, food and beverages. The proposed model includes three such antecedents: quality, communicating sustainability, and ethical behaviour. By including the interactive relations among these three antecedent factors, the analysis has revealed the impact of the three factors on consumers’ brand loyalty from a holistic perspective.

3. Methodology

3.1. Research Approach

The aim of the present research was to test and validate a conceptual model in view of the examination of the hypotheses developed from the literature review, with measurements related to a dataset collected from Thailand; thus, this study applies a quantitative approach [72]. The study involves the use of the survey method, specifically collecting data via questionnaire. This method is widely used, as it facilitates the collection of a large number of data from various sample groups [72,73].

3.2. Survey Design

In total, 26 items related to our hypotheses were prepared. Following a pilot test that was carried out with 12 volunteers, some modifications were made to the wording and the format before the main survey. The survey was conducted using a web-based survey service, SurveyMonkey, which allowed us to collect primary data from Bangkok, Thailand. The questions and corresponding options used in this research were carefully defined based on the key findings from the academic sources, following the model shown in Figure 1. The original survey questions were prepared in English before a bilingual individual translated them into Thai. A second bilingual individual, who had not seen the originals, subsequently back-translated the items into English before a third bilingual individual checked the translations. Throughout this process, any inconsistencies that emerged were discussed and resolved in order to finalise the questionnaire. This procedure was carried out with reference to [74,75] to ensure the cultural and language equivalency of the scales being used.

To validate the scales, the dataset was analysed via factor analysis, Cronbach’s alpha tests, and correlation analysis before the model was validated, with the dataset being utilised to examine the hypotheses using structural equation modelling (SEM).

3.3. Data Collection and Analysis

The questionnaire was distributed via electronic means (i.e., email and social network services) to a sample of 800 participants. The sample encompassed a wide range of age groups and both genders. All participants were residing in Bangkok, Thailand. The sample size was deemed suitable for conducting both the factor analysis and the subsequent SEM analysis [76]. SPSS version 26 was used for the data analysis, which was initially based on a descriptive analysis aimed at providing an overview of the dataset profile; this was followed by a relational analysis that incorporated factor analysis along with all the relevant observed variables. To test the reliability of the generated factors for further analysis, a Cronbach’s alpha test was conducted to validate each factor and variable [77,78]. Correlation analysis was also carried out to assess the multi-collinearity of the independent variables among the factors and confirm that they were suitable for the SEM analysis [79], which was conducted using AMOS version 26.

4. Results and Discussion

4.1. Data Profile

4.1.1. Descriptive Statistics

Table 2. Demographic profile.

Attribute		Frequency	Valid Percent	Cumulative Percent
Gender	Female	628	78.5	78.5
	Male	172	21.5	100.0
	Total	800	100.0
Age	10s	18	2.3	2.3
	20s	474	59.5	61.7
	30s	212	26.6	88.3
	40s	64	8.0	96.4
	50s	28	3.5	99.9
	60s	1	0.1	100.0
	Missing	3
	Total	800
Marriage status	Single	519	64.9	64.9
	Married	269	33.6	98.5
	Missing	12	1.5
	Total	800
Education	Secondary	38	4.8	4.8
	Undergraduate	596	74.9	79.6
	Postgraduate	148	18.6	98.2
	Others	14	1.8	100.0
	Missing	4
	Total	800
Occupation	Company worker	418	52.4	52.4
	Business owner	134	16.8	69.2
	Government	48	6.0	75.2
	Student	108	13.5	88.7
	Housewife	24	3.0	91.7
	Other	40	5.0	96.7
	Unemployed	26	3.3	100.0
	Missing	2
	Total	800
Income	Under 15K	180	22.6	22.6
	15K–25K	286	35.8	58.4
	25K–35K	150	18.8	77.2
	35K–45K	74	9.3	86.5
	45K–50K	32	4.0	90.5
	Over 50K	76	9.5	100.0
	Missing	2
	Total	800

shows the demographic profile of the sample of 800 individuals. Overall, women accounted for 78% of the total sample, which corresponds to the brand’s target demographic, while only around 22% of the sample were men. The average respondent age in our sample was 30.1 years old; moreover, 64.9% were single, while 33.6% were married or common-law married.

More detailed descriptive statistics for the dataset are shown in

Table 3. Descriptive statistics for the two sample groups, ‘Domestic brand’ and ‘International brand’.

	All Samples			Domestic Brand			International Brand
Observed Variables	N	Mean	Std. Dev.	N	Mean	Std. Dev.	N	Mean	Std. Dev.
BL1: I will purchase the same brand product again	800	2.63	1.01	400	2.70	0.99	400	2.56	1.04
BL2: I will purchase the same brand products even if other brands are being sold at discounted price.	800	2.78	0.99	400	2.91	0.95	400	2.65	1.01
BL3: I will recommend the brand I like to others	800	2.91	0.98	400	3.04	0.92	400	2.78	1.02
CS1: Attachment to the brand which has sustainability communication	800	3.13	0.91	400	2.94	0.85	400	3.31	0.93
CS2: Feel resonance to the brand with sustainable communication	800	3.00	0.91	400	2.89	0.91	400	3.11	0.91
CS3: Communicating sustainability is helpful to nurture my loyalty	800	3.19	0.84	400	3.18	0.84	400	3.19	0.84
QTY1: High quality of the products are important	800	3.36	1.08	400	3.80	0.93	400	2.93	1.05
QTY2: I prioritise the quality of ingredients and ethical sourcing, even the price is higher than others, when purchasing cosmetics	800	3.08	1.02	400	3.39	0.96	400	2.77	0.98
QTY3: I choose the brand which provides good quality of the products with clear discription and accountability	800	3.16	0.95	400	3.39	0.88	400	2.93	0.97
EB1: I pay attention to the brands’ societal activities	800	2.97	0.84	400	3.05	0.82	400	2.90	0.86
EB2: I choose brands which behave and act ethically	800	3.22	0.91	400	3.28	0.92	400	3.16	0.89
EB3: I choose brands which look after the earth	800	3.07	0.94	400	3.08	0.95	400	3.06	0.93

. This table divides the sample into two groups of 400—one for the ‘Domestic brand’ answers and one for the ‘International brand’ answers—and identifies the different tendencies for each observed variable. Moreover, the common method bias (CMB) was tested, and the result was 35.12%; this is less than 50% of the value specified by [80,81], indicating no CMB issues.

As discussed in the above literature review, the primary focus of this study is on the cognition and behaviour of the highly educated and relatively young generation in the Thai market, as this demographic represents the driving force behind purchasing behaviour. Therefore, a large proportion of respondents had undergraduate or postgraduate education (as demonstrated in Table 3), which indicates that the sample is skewed towards high education levels. This aspect needs to be carefully considered and properly reflected upon in the discussion while attempting to generalise the findings of the study. The attributes of the dataset will be discussed further below in the context of the study limitations and future research opportunities.

4.1.2. Confirmatory Factor Analysis and Reliability of the Scales

Confirmatory factor analysis was conducted for all samples. The results indicate the confirmation of the four specific factors, as shown in

Table 4. Factor analysis results.

Observed Variables	Component				Alpha
	1	2	3	4
BL1: I will purchase the same brand products again	0.820	0.157	0.181	0.114
BL2: I will purchase the same brand products even if other brands are selling at discounted price.	0.810	0.231	0.227	0.172	0.836
BL3: I will recommend the brand I like to others	0.722	0.286	0.330	0.174
CS1: Attachment to the brand which has sustainability communication	0.101	0.807	0.044	0.148
CS2: Feel resonance to the brand with sustainable communication	0.192	0.793	0.140	0.090	0.758
CS3: Communicating sustainability is helpful to nurture my loyalty	0.263	0.774	0.129	0.097
QTY1: High quality of the products are important	0.130	0.046	0.821	0.158
QTY2: I prioritise the quality of ingredients when purchasing cosmetics	0.246	0.110	0.767	0.153	0.769
QTY3: I choose the brand which provides good quality of the products	0.255	0.174	0.708	0.171
EB1: I pay attention to the brands’ societal activities	0.174	0.099	0.049	0.824
EB2: I choose brands which behave and act ethically	0.035	0.110	0.175	0.764	0.745
EB3: I choose brands which look after the earth	0.190	0.132	0.270	0.742
Sums of Squared Loadings	2.175	2.123	2.101	2.003
% of Variance	18.123	17.695	17.512	16.688
Cumulative %	18.123	35.818	53.330	70.018

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Rotation converged in 5 iterations.

. The Cronbach’s alphas for the four factors were 0.836, 0.758, 0.769, and 0.745, respectively; with all alpha values greater than 0.7, the validity and reliability of the factors were confirmed [82].

To investigate the difference in consumer perspectives between the two brands, a non-parametric test was conducted to check the statistical significance between the two sample groups, the results of which are presented in

Table 5. Non-parametric test results.

	Maker	N	Mean Rank	Sum of Ranks
Purchase Intention	Domestic brand	400	433.92	173,570
	International brand	400	367.08	146,831
	Total	800
Quality	Domestic brand	400	364.45	145,778
	International brand	400	436.56	174,622
	Total	800
Attachment to Brand	Domestic brand	400	491.86	196,744
	International brand	400	309.14	123,656
	Total	800
Ethical Behaviour	Domestic brand	400	419.30	167,719
	International brand	400	381.70	152,682
	Total	800
Test Statistics	Purchase Intention	Quality	Attachment to Brand	Ethical Behaviour
Mann−Whitney U	66,630.500	65,578.000	43,456.000	72,481.500
Wilcoxon W	146,830.500	145,778.000	123,656.000	152,681.500
Z	−4.108	−4.445	−11.251	−2.317
Asymp. Sig. (2−tailed)	0.000	0.000	0.000	0.020

.

As Table 5 shows, the non-parametric test results indicate that the null hypothesis (‘there is no difference between the two groups’) was rejected for the four main factors: brand loyalty, quality, communicating sustainability, and ethical behaviour. It can therefore be stated that significant differences in consumer perspectives exist between the two groups in terms of all four of the latent factors.

4.1.3. Convergent and Discriminant Validity Tests

Once the process of ascertaining how each item can explain its own construct was completed, the next examination was conducted to identify whether the constructs are valid and reliable. The composite reliabilities (CRs) and average variance extracted (AVE) of the constructs were computed by statistical procedure [83]. These were estimated and are presented in

Table 6. Convergent and discriminant validity.

	N	Mean	SD	CA	CR	AVE	Brand Loyalty	Quality	Communicating Sustainability	Ethical Behaviour
Brand loyalty	800	2.775	0.862	0.836	0.828	0.616	0.785
Quality	800	3.103	0.728	0.758	0.834	0.627	0.564 **	0.792
Communicating Sustainability	800	3.202	0.842	0.769	0.810	0.588	0.504 **	0.338 **	0.767
Ethical Behaviour	800	3.088	0.729	0.745	0.820	0.604	0.413 **	0.446 **	0.319 **	0.777

Note 1: Values bold on the main diagonal are the square rooted of AVEs; SD is standard deviation; CA is Cronbach alpha; CR is Composite reliability; AVE is average variance standard. Note 2: ** means p < 0.1.

. Most of the examined values are greater than the accepted lowest values of CR and AVE, which are 0.7 [84] and 0.5 [78]. As Table 6 shows, all values have met the relevant requirements, and the overall outcome implies that the constructs are reliable, consistent, and valid [83].

Next, a discriminant validity test was conducted. If the square root of the AVE of each construct, or the average variance (AV), is found to be greater than the Pearson correlation coefficient of that construct with other constructs, then it can be concluded that discriminant validity is confirmed. We computed AVs and Pearson correlation coefficients as demonstrated in Table 6. It was confirmed that the Cronbach’s alpha of each construct also indicates that each construct is consistent, as the values of the Cronbach’s alpha of each construct are greater than the lowest acceptable value of 0.6 [85]. Constructs so identified are said to be consistent [83]. Correlation analysis is useful for detecting the existence of a covariant among relevant factors that constitute the SEM. As high correlation indicates multicollinearity among variables, the correlation coefficients would ideally not be higher than 0.7 [86]. As shown in Table 6, the factors are confirmed to be suitable for SEM analysis.

4.2. Structural Equation Modelling

Using the four factors confirmed via the factor analysis, SEM analysis was conducted on three datasets: ‘All’, ‘Domestic brand’, and ‘International brand’. Figure 1 presents the SEM results for all samples along with the statistical coefficients of each relational path. As one of the best-fit measures for the research model, the χ²/degrees of freedom was 2.187 (χ² = 105.004, df = 48), which met the recommended level of <5.0 [85]. Similarly, the other indicators all demonstrated a good fit (GFI = 0.979, AGFI = 0.966, and CFI = 0.984) and were all high enough to meet the recommended level of >0.90. Meanwhile, the RMSEA (0.039) also met the recommended level of <0.10 [85], indicating that the analytical model was reliable and suitable for the dataset. Moreover, the R2 value for brand loyalty was relatively high at 0.638.

The SEM results show that the coefficient of ‘quality to brand loyalty’ was 0.493 ***, while that of ‘communicating sustainability to brand loyalty’ was 0.393 ***. However, no significant coefficient was found for the ‘ethical behaviour to brand loyalty’ path. All three of the main factors demonstrated interrelations with significant covariances.

In view of the above, the results of the hypothesis testing can be summarised as follows:

H1 is supported: Quality has a significant impact on consumers’ brand loyalty towards cosmetic products.

H2 is supported: Communicating sustainability has a significant impact on consumers’ brand loyalty towards cosmetic products.

H3 is rejected: The ethical behaviour of the producers has no significant impact on consumers’ brand loyalty towards cosmetic products.

H4 is supported: The three main factors for consumer loyalty towards a cosmetics brand are all interrelated.

In particular, as Figure 1 demonstrates, the results of this analysis with respect to H3 have very interesting implications (i.e., that the ethical behaviour of companies has no significant impact on brand loyalty), providing evidence that young Thai consumers, especially those with higher levels of education, do not perceive the CSR activities of cosmetic suppliers as an important factor in brand loyalty. These results may indicate that the younger, more educated generation in Thailand, being located in an emerging market, is in a transition period in terms of market and consumer behaviour, and accordingly that their perceptions of and behaviour towards CSR may differ from those of consumers in developed countries.

4.3. Comparative Analysis of the Two Brands

4.3.1. Domestic Brand

The SEM model for the ‘Domestic brand’ demonstrated a good fit with a χ²/degrees of freedom of 2.131 (χ² = 102.309, df = 48), while the other indices were also above the recommended level of >0.90 (GFI = 0.961, AGFI = 0.937 and CFI = 0.965) and <0.10 (RMSEA = 0.053) [85]. Moreover, the brand loyalty R2 value was relatively high at 0.679.

The coefficients of all pathways from two of the factors (‘quality’ and ‘communicating sustainability’) to ‘brand loyalty’ were 0.737 *** and 0.206 **, respectively. Meanwhile, the path between ‘ethical behaviour’ and ‘brand loyalty’ demonstrated no significant relationship. The covariances between the three factors indicated significant interrelations, suggesting that while the direct relationship between ethical behaviour and brand loyalty is not significant, all three of the factors could have an indirect impact via influence on the other factors.

4.3.2. International Brand

The SEM model for the ‘International brand’ also indicated good fit levels, with a χ²/degrees of freedom of 1.852 (χ² = 88.938, df = 48), while the other indicators met the recommended level of >0.90 (GFI = 0.965, AGFI = 0.943 and CFI = 0.980) and <0.10 (RMSEA = 0.053) [85]. Moreover, the R2 value for brand loyalty was relatively high at 0.635.

The coefficients of all three paths between ‘quality’, ‘communicating sustainability’, ‘ethical behaviour’ and ‘brand loyalty’ were all found to be significant, while all covariances among the three factors regarding the relevant hypotheses also demonstrated significant relationships, suggesting that all these factors are interrelated and capable of influencing ‘brand loyalty’.

Table 7. Path coefficients among the variables.

Path			All Samples		Domestic Brand		International Brand
To		From	Path Coefficient	p	Path Coefficient	p	Path Coefficient	p
Brand Loyalty	<---	Quality	0.493	***	0.737	***	0.333	***
Brand Loyalty	<---	Communicating Sustainability	0.393	***	0.206	**	0.466	***
Brand Loyalty	<---	Ethical Behaviour	0.072	0.13	−0.044	0.610	0.124	*
BL1	<---	Brand Loyalty	0.831	1 fix	0.827	1 fix	0.832	1 fix
BL2	<---	Brand Loyalty	0.709	***	0.661	***	0.754	***
BL3	<---	Brand Loyalty	0.844	***	0.812	***	0.861	***
QTY1	<---	Quality	0.744	1 fix	0.770	1 fix	0.774	1 fix
QTY2	<---	Quality	0.683	***	0.624	***	0.709	***
QTY3	<---	Quality	0.717	***	0.670	***	0.729	***
CS1	<---	Communicating Sustainability	0.662	1 fix	0.465	1 fix	0.733	1 fix
CS2	<---	Communicating Sustainability	0.740	***	0.683	***	0.737	***
CS3	<---	Communicating Sustainability	0.778	***	0.757	***	0.748	***
EB1	<---	Ethical Behaviour	0.692	1 fix	0.690	1 fix	0.696	1 fix
EB2	<---	Ethical Behaviour	0.613	***	0.577	***	0.643	***
EB3	<---	Ethical Behaviour	0.798	***	0.768	***	0.825	***
Quality	<-->	Ethical Behaviour	0.589	***	0.677	***	0.551	***
Quality	<-->	Communicating Sustainability	0.440	***	0.475	***	0.640	***
Communicating Sustainability	<-->	Ethical Behaviour	0.425	***	0.451	***	0.419	***
Fit Indexes		Chi-square	105.004		102.309		88.938
		df	48		48		48
		C/D	2.188		2.131		1.853
		p	0.000		0.000		0.000
		GFI	0.979		0.961		0.965
		AGFI	0.966		0.937		0.943
		CFI	0.984		0.965		0.980
		RMSEA	0.039		0.053		0.046
R2		Brand Loyalty	0.638		0.679		0.635
		BL1	0.691		0.683		0.693
		BL2	0.503		0.437		0.569
		BL3	0.712		0.660		0.742
		QTY1	0.553		0.593		0.599
		QTY2	0.466		0.390		0.503
		QTY3	0.514		0.449		0.531
		CS1	0.438		0.216		0.537
		CS2	0.547		0.466		0.544
		CS3	0.606		0.573		0.559
		EB1	0.478		0.477		0.485
		EB2	0.375		0.333		0.414
		EB3	0.636		0.590		0.680

Note 1: df: degree of freedom, C/D: Chi-square/df, p: provability, GFI: Goodness of fit index, AGFI: Adjusted goodness of fit index, CFI: Comparative fit index, RMSEA: Root mean square error of approximation. Note 2: Squared multiple correlations (SMC) in SPSS AMOS was used as R2. Note 3: *** means p < 0.001; ** means p < 0.01; * means p < 0.05 Note 4: 1 fix means regression weight is fixed as 1.

presents the results for all three models.

4.3.3. Overview of the Comparative Analysis

The Thai consumers’ perspectives on the ‘Domestic brand’ in terms of ‘quality’ and ‘brand loyalty’ were found to have the highest significance (0.737 ***), while ‘communicating sustainability’ and ‘brand loyalty’ demonstrated a weaker correlation (0.206 **). In terms of the ‘International brand’, all three factors had a significant impact on ‘brand loyalty’, while the impact of ‘quality’ was weaker than that for the ‘Domestic brand’.

In the ‘All’ sample, two of the three latent factors (i.e., not ethical behaviour) were found to have a significant effect on brand loyalty, while no such significant relationship was found for ethical behaviour. This phenomenon can be interpreted to suggest that consumers in Thailand, as an emerging country, already pay attention to marketing communications about the quality and sustainability of cosmetics and judge these factors as triggers for purchase and repeat business. The three observables of communicating sustainability, or CS (‘Attachment to brand with CS’; ‘Feeling resonance with CS’; ‘CS is helpful for nurturing my loyalty’) are all significant components of the latent factor CS. On the other hand, ethical behaviour (EB), which consists of three observables (‘Social activities’; ‘Acting ethically’; ‘Looking after the earth’), has not yet been grasped and evaluated with a clear awareness or recognised as a reference factor in product purchasing (see Figure 1 for all samples.) Further details are presented in Table 7.

In light of Table 7, the significance of ethical behaviour is found to differ between domestic and international brands. In other words, the relationship between domestic brands and H4 is not supported. Thai consumers may change the focus of their purchase decision depending on the product’s country of origin. This suggests that consumers in emerging countries will pay closer attention to manufacturers’ ethical behaviour in the future and allow this information to influence their brand loyalty.

Moreover, H4 ‘covariance relationship between the three latent factors’, was found to be significant; thus, although no direct relationship between ethical behaviour and brand loyalty was identified, it is possible that ethical behaviour influences brand loyalty via the other two antecedents.

To triangulate this finding, the four factors for the two brands were examined based on the scores for the confirmed factors (Figure 2).

As Figure 2 shows, significantly different perspectives can be observed between the two brands. The most substantial difference was found for ‘communicating sustainability’, which indicates that Thai consumers are more responsive to the domestic brands’ marketing strategies for communicating sustainability than those of the international brands.

The finding of this study could fill the gap in the existing research. For example, [87] conducted an experiment to test whether and how CSR marketing influences consumer behaviour and found that the influence increases when the psychological distance is close. On the other hand, [88] attempted to examine the main and interaction effects of CSR involvement, brand naming and pricing on consumer purchase intentions using a 2 × 2 × 2 experimental design. In the high price range, CSR involvement had a strong effect on purchase intentions for products with foreign brand names; in the low price range, CSR involvement had a significant effect regardless of brand name. These authors concluded that there is no significant effect of CSR involvement regardless of the brand name. Among more recent studies, for example, [89] examined the mediating effect of CSR factors on Gen-Y consumer behaviour and found a direct relationship between perceptions of value compatibility and consumer perceptions of CSR.

Thus, while several studies using Thai consumer datasets have examined the impact of CSR-based marketing on consumers’ willingness to purchase, few studies have discussed the differences in consumer behaviour due to CSR. Among the valuable existing studies, for example, [7] examine the interaction between CSR and branding in the context of an emerging economy, using data from a survey of the attitudes of residents of the Lima metropolitan area in Peru, and conducted an analysis using the stimulus-organisation-response (SOR) model. The study argues that CSR effectively acts as a stimulus for consumers to identify and associate with a brand. However, these authors also note that given the impact of emerging economies on the global economy, alongside the magnitude of the impact of CSR initiatives on consumer behaviour in those countries, there is far too little research on the subject.

5. Conclusions, Implications and Limitations

5.1. Theoretical Contributions

This study has revealed and validated three antecedent factors, demonstrating that they are useful in examining consumers’ loyalty towards branded cosmetic products. Among the three factors, the ‘ethical behaviour of the producers’ factor was found to have a non-significant impact (for all samples and ‘Domestic brand’) and the weakest impact on brand loyalty for ‘International brand’ products. Meanwhile, ‘quality’ had the most significant impact on brand loyalty, especially for the ‘Domestic brand’, indicating that domestic brands should focus on quality to develop customer loyalty.

Interesting outcomes were found for the ‘communicating sustainability’ and ‘ethical behaviour of the producers’ factors, with the former having a significant impact on brand loyalty across all samples and the latter demonstrating no significant impact and having the weakest impact on brand loyalty. As numerous scholars have indicated, communicating sustainability and ethical behaviour are important elements of devising marketing strategies to attract customers [90,91,92]. Moreover, ensuring brand loyalty is evidently another target for firms aiming to be more competitive and sustainable [35,93,94].

Here, while we analysed the impact of the aforementioned factors within the context of business sustainability strategies, the present conclusions must be further verified using a larger dataset in order to develop robust implications for both theory and practice. Given that, due to the nature of emerging markets, the current snapshot of consumer perspectives and behaviours in these markets could well change over time, it will be crucial to conduct further research that extends beyond the established international knowledge and experience of consumer behaviour [95,96].

Scale development and validation within an emerging market context have been confirmed in this study, while consumer perceptions and behaviours within a Thai market context were found to exhibit different characteristics from the oft-discussed Western markets. Therefore, the authors acknowledge that the present research findings will help to identify the changes between emerging markets and can also help to reveal the differences between emerging and developed markets.

However, our finding that the impact of sustainable communication and ethical values on consumer loyalty is non-significant stands in stark contrast to the findings of previous studies working within different contexts. Further research into this discrepancy is undoubtedly required. Another issue that merits further investigation is as follows: since the data for this study were largely collected from a sample of highly educated people (university degree or above) living in the capital city of Bangkok, the findings of a non-significant influence of CSR and ethical behaviour of producers on the purchase intentions of Thailand’s highly educated, urban-dwelling Gen-Y cohort should be explored in more depth.

5.2. Practical Contribution

This study makes several novel practical contributions. The first is that it was based on a relatively large sample of well-educated young consumers (93.5% of the total sample are university graduates or have attained postgraduate qualifications), making it well-suited to assessing the younger generation’s perspectives and attitudes.

The first significant output from the study is that it demonstrates that young Thai consumers tend not to pay attention to the ethical behaviour displayed by businesses when they make purchasing decisions in the cosmetics market. This brings into focus an issue that has not been substantially addressed by previous research into buying behaviour in emerging markets generally and among younger Thai consumers specifically: in short, at present, these consumers do not give much thought to the ethical behaviour of cosmetic manufacturers, a finding that should be taken into account when formulating marketing strategies for cosmetic companies.

On the other hand, these results also highlight the need to consider differences in consumer attitudes depending on whether the brand is international or domestic, so as to determine which channels can be used to gain their support and how current consumer behaviour is likely to change over time. In particular, the results clearly suggest that we should carefully monitor the ways in which current consumer behaviour changes over time and reflect this in our business strategy.

Of particular relevance is the development of a practical measure of the determinants of consumer purchasing behaviour in emerging markets. The results of an empirical investigation using this measure revealed that consumer awareness and behaviour differ depending on the brand’s country of origin. We believe that the results of this study provide a valuable basis for the development of future corporate strategies for communicating with consumers about sustainability.

5.3. Limitations

As discussed at the beginning of this study, the analysis is based on a sample of young people, most of whom are highly educated, who represent the driving force behind Thailand’s economy owing to their socio-economic situation. We thus cannot simply generalise these findings to Thai consumers in general. The authors are well aware that results based on such a distinct dataset must be tested in future on a broader dataset and discussed from a higher perspective.

As noted, the results obtained here stand in contrast to the findings of international research, which has shown that younger generations are more likely than other generations to engage in ethical business practices and to pay attention to the CSR strategies of product and service providers in their purchasing behaviour. In other words, H3 is not supported, which may suggest that the preferences of young Thai consumers are still undergoing a transition due to their status as participants in an emerging market. In light of the need to promote sustainable development and responsible manufacturing and consumption practices, further related research, accumulation of knowledge and policy implications should be pursued.

5.4. Recommendations and Future Work

As noted above, the findings of this study should be re-examined in more depth; however, the present discussion with its basis in a Thai dataset will unquestionably shed some light on the current research agenda. Nonetheless, further and more comprehensive investigations into consumers’ perspectives on and attitudes towards different brands should be conducted in the future. The notion is that a good brand image that incorporates ethical values and communicates sustainable efforts will go some way towards achieving consumer loyalty. It is the responsibility of social scientists to observe and track consumers’ changing perspectives and attitudes toward sustainable and ethical business strategies, as this will aid them to evaluate and recommend how effective marketing plans can be formulated within competitive markets.

In future studies, the scales and measurements developed herein will be validated and tested for other industrial sectors (e.g., food, apparel, and other commodities) and in different markets with varied values and histories in order to expand our contribution to both theory and practice.