Influencing millennials to embrace sustainable fashion in an emerging market: a modified brand avoidance model perspective

Abstract

Purpose

This study applies the modified brand avoidance model to examine factors that influence sustainable fashion avoidance behaviour among millennial shoppers in South Africa.

Design/methodology/approach

A positivistic approach and a web-based online survey were employed to collect cross-sectional data from 423 millennial fashion shoppers. Standard multiple regression analysis was used to test proposed hypotheses.

Findings

Unmet expectations, materialism and symbolic incongruence emerged as major predictors of millennials' intention to avoid sustainable fashion. Sustainable fashion avoidance intention was found to have a positive effect on sustainable fashion avoidance behaviour.

Research limitations/implications

This study relied on self-reported data collected from millennial shoppers. Future studies may improve the generalizability of this study's results by conducting a comparative study with other cohorts such as baby boomers and Generation X who espouse different shopping values. Future studies may benefit from the use of longitudinal data in order to understand how millennial shoppers relate to sustainable fashion as it evolves.

Practical implications

The results of this study suggest the importance of developing value propositions that align sustainable fashion with cultural, personality and symbolic cues valued by millennial shoppers. Consumer education on the benefits of sustainable fashion is recommended as a long-term behavioural change strategy.

Social implications

The purchase behaviour of sustainable fashion should be encouraged as it enhances environmental sustainability including safeguarding the livelihoods of future generations.

Originality/value

This study contributes to literature on sustainable fashion avoidance behaviour. This is one of the pioneering studies to empirically examine the influence of unmet expectations, symbolic incongruence and ideological incompatibility in the context of an emerging market, such as South Africa.

Citation

Muposhi, A. and Chuchu, T. (2024), "Influencing millennials to embrace sustainable fashion in an emerging market: a modified brand avoidance model perspective", Journal of Fashion Marketing and Management, Vol. 28 No. 4, pp. 738-758. https://doi.org/10.1108/JFMM-07-2021-0169

Introduction

The fashion industry is one of the major contributors to environmental pollution (Woodside and Fine, 2019; Peters et al., 2021). Globally, carbon emissions from the fashion industry are estimated to be 1.7 billion tonnes (Loetscher, 2017). It is also projected that at least 8,000 chemicals are required to convert raw materials into fashion fabrics (Woodside and Fine, 2019). The environmental harm linked to the fashion industry is worsened by the growth in popularity of fast fashion. Fast fashion is widely regarded as the antithesis to the goal of environmental sustainability (Choi and Cai, 2018; McNeill and Venter, 2019). This is so because fast fashion is characterized by the marketing of mass produced and trending apparel, often with fast disposal rates (Gazzola et al., 2020). As a business model, fast fashion is considered unsustainable due to labour exploitation (Pal and Gander, 2018), massive consumption of energy and water (Wang et al., 2019), rapid disposal of apparel waste in landfills (Niinimaki et al., 2020) and high discharge of hazardous chemicals and greenhouse gases (Bentahar and Benzidia, 2018). The 2013 Rana Plaza accident in Savar, Bangladesh, is commonly cited as an example of unsustainable practices in the fashion industry (Peters et al., 2021).

Developing economies suffer the most from the ills of fast fashion, such as the exploitation of labour, environmental pollution and dumping of cheap poor-quality apparel, yet the adoption of sustainable fashion remains low (Cimatti et al., 2017; Gazzola et al., 2020). South Africa, as one of the major economies in Africa and a member of BRICS (Brazil, Russia, India, China and South Africa) trading bloc, has not been spared from the excesses of fast fashion. The widespread influx of cheap, substandard imported apparel from China has been blamed for almost collapsing South Africa's textile industry (Wasserman, 2012). In an attempt to address the adverse environmental impact of fast fashion, sustainable fashion, also known as slow fashion, has emerged as an environmentally friendly alternative (Bentahar and Benzidia, 2018). Sustainable fashion is defined as apparel that is manufactured using less water and that can be reused, composted or recycled (Choi and Cai, 2018).

The rise in environmental concern, along with the need to curtail the negative externalities attributed to fast fashion, was expected to stimulate favourable consumer perceptions towards sustainable fashion (Khandual and Pradhan, 2019; Ripple et al., 2019). Expecting a transition towards sustainable fashion by shoppers, research efforts were directed towards understanding the drivers of fast fashion avoidance (e.g. Kim et al., 2013; Knittel et al., 2016; Yoon et al., 2020). However, it seems the anticipated sustainable fashion adoption bandwagon has not yet materialized in emerging markets such as South Africa. Although South African shoppers report that they support the call to shift from fast fashion to sustainable fashion, the actual behaviour of buying environmentally friendly apparel remains low (SASTAC, 2014; May, 2019). This has been referred to as the sustainable fashion paradox (Bernardes et al., 2018;). Of major concern to marketers of sustainable apparel in emerging markets is that millennial shoppers, who are regarded as the most influential cohort, based on purchasing power, fashion consciousness and environmental knowledge, remain hesitant to buy sustainable fashion (Vuong and Nguyen, 2018; Samala and Singh, 2019; Pencarelli et al., 2020). In South Africa, despite concerted efforts by organizations such as SA Fashion Week, Cotton On, Woolworths and fashion designers (Akina, Pichulik, Sitting Pretty and Lunar) to promote sustainable fashion, May (2019) notes that demand remains low and continues to be skewed towards fast fashion. It is against this background that this study responds to the call by Rolling and Sadachar (2018), as well as Pencarelli et al. (2020), to understand the motivations and apprehensions of millennial shoppers towards sustainable fashion.

Research objectives

Sustainable fashion has generated significant research interest, with previous studies focusing on consumer perceptions (Sung and Woo, 2019), sustainable fashion controversy (Greco and De Cock, 2021) and sustainable fashion purchase behaviour (Bernardes et al., 2018; Iran et al., 2019; Peters et al., 2021). Favourable attitudes of millennials towards sustainable fashion were found not to translate into actual purchase behaviour (Bernardes et al., 2018; Vuong and Nguyen, 2018). Millennial shoppers were also found not to prioritize the concept of sustainability when making purchase decisions (Iran et al., 2019). While previous studies (e.g. Kim et al., 2013; Knittel et al., 2016; Yoon et al., 2020) focused on understanding drivers of fast fashion avoidance, empirical evidence suggests that millennials are actually preferring fast fashion as opposed to sustainable fashion (Bernardes et al., 2018; Rolling and Sadachar, 2018; Pencarelli et al., 2020). As far as we can establish, there is little empirical evidence on why millennials shoppers avoid buying sustainable fashion. Thus, this study sets out to understand, using the modified brand avoidance model (BAM), the underlying factors that influence millennial shoppers to avoid buying sustainable fashion.

Of particular interest to this study is understanding why millennial South African fashion shoppers, who are known to be concerned with the significant loss of employment in the textile industry due to the influx of fast fashion (SASTAC, 2014) and the high levels of environmental pollution (O'Brien and Thondhlana, 2019), are reluctant to embrace sustainable fashion. Given this background, the research question central to this study is: Why do millennial shoppers prefer to buy fast fashion instead of sustainable fashion in South Africa? This study employs the BAM (Lee et al., 2009) and the personality trait theory (Allport, 1961) in an attempt to address two specific objectives: (1) To examine antecedents of sustainable fashion avoidance intention in South Africa and (2) to understand the relationship between sustainable fashion avoidance intention and sustainable fashion avoidance behaviour. The remainder of this paper is structured as follows: the next section reviews literature on sustainable fashion. Thereafter, the conceptual framework and hypotheses are developed. Research methodology, data analysis and interpretation of results then follow. The last sections provide implications, conclusion, limitations and suggestions for further research.

Review of literature

Slowing fast fashion – sustainable fashion drivers and counter forces

The fashion industry is increasingly under pressure to embrace sustainability principles in order to enhance environmental citizenship status (Jiang et al., 2018; Peters et al., 2021). In this regard, a transition from fast fashion to sustainable fashion is imperative (Kozlowski et al., 2018). Fast fashion is perceived as an embodiment of unsustainability due to high water and energy consumption, environmental pollution and unfair labour practices (Wang et al., 2019; Niinimaki et al., 2020). In order to sustain a low-cost strategy, fast-fashion companies often situate their manufacturing plants in developing economies with cheap labour, minimum levels of environmental concern and relaxed environmental regulations (Napier and Sanguineti, 2018; Peters et al., 2021). Sustainable fashion, which is also known as slow fashion, green fashion or eco-fashion, refers to apparel manufactured using raw materials and manufacturing practices with minimum harm to the natural environment (Niinimaki et al., 2020). In practice, the manufacturing of sustainable fashion involves the use of sustainable sourcing of raw materials, such as organic cotton or recycled fabrics, sustainable design (design for recyclability and eco-friendly dyeing methods), fair labour practices and the use of eco-friendly packaging (Moorhouse and Moorhouse, 2017; Chiu et al., 2018).

It is important to state that there are competing value propositions used by marketers of fast and sustainable fashion. Fast fashion is positioned as trendy, cost effective, including ability to enhance instant gratification (; Yoon et al., 2020). The attribute of trendiness resonates with the personality traits of millennial shoppers of trend setting, exclusivity and materialism, especially for fast fashion apparel that mimic luxury brands (Skarmeas and Leonidou, 2013; Ladhari et al., 2019). Fast-fashion loyalists perceive it as ideal since it captures the essence of fashion, that is, trendiness (Murphy and Schlegelmilch, 2013). The low-cost strategy is a major pull factor as fast fashion is considered more affordable as compared to sustainable fashion, especially in developing economies (Armstrong et al., 2015; Ciasullo et al., 2017). The downside of fast fashion emanates from its association with environmental harm and temporality – the speed at which it becomes outdated (McNeill and Venter, 2019). The short life cycle of fast fashion is also blamed for nourishing a materialistic fashion consumption culture which is considered unsustainable (Skarmeas and Leonidou, 2013).

As the recommended alternative, sustainable fashion employs environmental consciousness and longevity as unique selling propositions (Moorhouse and Moorhouse, 2017). Sustainable fashion seeks to tap into the market potential of a growing segment of environmentally conscious shoppers that is willing to curb the scourge of climate change through responsible purchase behaviour (Gazzola et al., 2020). The attribute of longevity addresses the problem of rapid disposal associated with fast fashion as it promotes long-term use of apparel and reusability (McNeill and Moore, 2015). By encouraging reuse and recyclability, sustainable fashion promotes a circular economy which reduces environmental harm (Moorhouse and Moorhouse, 2017). By investing in sustainable fashion, companies potentially gain a positive corporate image due to the associated good environmental stewardship status (Chan and Wong, 2012; Wang et al., 2019).

Sustainable fashion has inherent controversies worth noting. For instance, there are many definitions of sustainability (Evans and Peirson-Smith, 2018; Ritch, 2021). This, according to Ritch (2021), presents a daunting challenge in terms of harmonizing its practice. The concept of sustainable fashion is anchored on the triple bottom line principle that emphasizes the need to maintain an intricate balance of environmental, economic and social goals (Chandran and Bhattacharya, 2019). Critics of the sustainable fashion movement perceive it as a ploy to maximize profit by using the environment as a selling proposition (McNeill and Moore, 2015; Brandao and Costa, 2021). Moreover, greenwashing concerns have been raised, defined as the act of using unsubstantiated environmental claims to market fashion products (Fernando et al., 2014; Mahsud et al., 2018).

Critics of sustainable fashion are also concerned with the high premium price which is beyond the reach of many consumers (Ritch, 2015; Brandao and Costa, 2021). For instance, previous studies found that consumers are only willing to pay an average of 10–20 per cent premium on sustainable fashion (Žurga and Forte, 2014; Ciasullo et al., 2017). Sustainable fashion is also criticized for attempting to standardize fashion by neglecting the notion of trendiness which is regarded as a central feature of fashion (Wang et al., 2019). The sustainable fashion movement has also been criticized for failing to consider the heterogeneity of consumer culture during the eco-design process (Brandao and Costa, 2021; Chakraborty and Sadachar, 2022). In practice, Caniato et al. (2012) also note the challenge confronted by fashion designers as some major suppliers of sustainable raw materials may not be willing to disclose critical sustainability-related data.

Millennials and sustainable fashion consumption

The generational cohort theory (GCT) is applied in marketing studies as a framework for consumer profiling (Eastman and Liu, 2012;). The GCT posits that individuals who were born during a particular period share similar life experiences that predispose them to respond to marketing stimuli in a similar manner (Eastman and Liu, 2012). Of interest to this study is the millennial or Generation Y cohort which is represented by individuals who were born between 1981 and 2000 (Bento et al., 2018). Millennials are distinguished from other cohorts by personality traits of independence, technologically savvy, inner-directedness and high level of purchasing power (Eastman and Liu, 2012). Although accurate statistics on the market size and market potential are not available, it is estimated that millennials contribute to more than 50 per cent of retail sales in South Africa (Duh and Struwig, 2015). Millennials are also known to be fashion conscious which make them a prime target for fashion retailers (). Millennials are also reported to possess higher levels of environmental concern by virtue of them being born and grown up in an environment of heightened environmental consciousness (Lu et al., 2013; Sunf and Woo, 2019). This explains why millennials are expected to embrace environmentally friendly behaviours such as the purchase of sustainable fashion (Rolling and Sadachar, 2018; Pencarelli et al., 2020).

Extant literature, however, suggests that millennials are not living up to their expected role of driving sustainable consumption (Bernardes et al., 2018; Pencarelli et al., 2020). According to Bernardes et al. (2018), translating millennials' environmental concern into actual purchase behaviour of sustainable products remains a challenge. This points to the existence of constraining factors that explain this behavioural inconsistency. In this regard, the anti-fashion consumption literature could offer a plausible explanation to the mismatch between millennials' environmental concern and low preference for sustainable fashion. In the extant literature, the term “brand avoidance” is commonly used to refer to incidences of brand rejection (Lee et al., 2009) and reduced brand consumption (Yoon et al., 2020). The drivers of brand avoidance include moral beliefs, dissatisfaction, brand image incongruence and unmet consumer expectations (Lee et al., 2009; Yoon et al., 2020). Against this background, this study examines whether brand avoidance dimensions and the personality trait of materialism may explain millennials' apathetic behaviour towards sustainable fashion.

Theoretical framework and hypotheses development

It is established in the marketing literature that an understanding of consumer needs is equally important as knowing what they do not prefer (Schiffman et al., 2010). This study applies Lee et al.'s (2009) BAM and the personality trait theory (Allport, 1961) to answer the research question: Why are millennial fashion shoppers in South Africa reluctant to buy sustainable fashion despite reporting higher levels of environmental concern? The BAM was applied in previous studies to understand consumer perceptions towards fast fashion (Kim et al., 2013; Knittel et al., 2016; Yoon et al., 2020). Brand avoidance is a form of consumerism movement in which consumers reject buying certain brands (Lee et al., 2009). The BAM identifies ideological incompatibility, symbolic incongruence and unmet expectations as factors that influence brand avoidance intentions (Lee et al., 2009). This study extends the BAM by including the personality trait of materialism to examine the extent to which the compulsive behaviour orientation of millennials influences sustainable fashion avoidance intention. Materialism is important to this study because millennials were found to have higher materialistic tendencies, compared to other consumer cohorts (Eastman and Liu, 2012; Colucci and Scarpi, 2013). In accordance with the theory of planned behaviour's proposition that intention is the most immediate precursor of behaviour (Ajzen, 1991), this study further examines the relationship between sustainable fashion avoidance intention and sustainable fashion avoidance behaviour.

Ideological incompatibility and sustainable fashion avoidance intention

Ideological incompatibility refers to a set of socio-economic and political world views that influence purchase behaviour (Lee et al., 2009). Foreignness and irresponsibility are identified as the main dimensions of ideological incompatibility (Lee et al., 2009). Sustainable fashion is criticized for attempting to standardize fashion in a manner that negates established cultural values (Kozlowski et al., 2019). The perception of sustainable fashion among millennials is adversely affected by the plurality of its value proposition (Vuong and Nguyen, 2018, Ritch, 2021). Proponents of sustainable fashion cherish it as a moral response by the fashion industry to self-correct, following decades of unchecked environment harm (Biswas et al., 2018; Kozlowski et al., 2019). On the other hand, the irresponsibility view emanates from the perception that a focus on sustainable fashion is a ploy to maximize profits by charging a premium price for greening fashion (Gazzola et al., 2020). For some, sustainable fashion is regarded as another shade of green capitalism (Žurga and Forte, 2014; Ciasullo et al., 2017). Previous studies on millennial shoppers' attitudes towards sustainable fashion report mixed results (Rolling and Sadachar, 2018; Heo and Muralidharan, 2019; Pencarelli et al., 2020). Based on findings from previous studies, it is reasonable to speculate that millennials may seek to avoid sustainable fashion due to ideological incompatibility and the green capitalistic tag affixed to it. Thus, it is hypothesized that:

H1.

Ideological incompatibility positively influences millennial shoppers' sustainable fashion avoidance intention.

Unmet expectation and sustainable fashion avoidance intention

Consumers are known to patronize brands that satisfy their needs and avoid those that fail to meet their expectations (Schiffman et al., 2010). Unmet consumer expectations are triggered by poor brand performance and post-purchase dissonance (Lee et al., 2009). Previous studies found that millennials perceive sustainable fashion to be lacking in trendiness, style and hedonism (Colucci and Scarpi, 2013; Samala and Singh, 2019). For millennial shoppers, fashion is construed as a way of self-expression through the acquisition of trending styles (McNeill and Moore, 2015). Sustainable fashion practices that emphasize reusability, remanufacturing, recyclability and collaborative consumption, along with the high premium price charged, are identified as the major sources of the quality and value for money gaps perceived by consumers (Gazzola et al., 2020). Previous studies (Fletcher, 2008; Shen et al., 2014; Newman et al., 2014) found that consumers are not prepared to forgo the utilitarian product benefits, such as quality and performance, for the sake of buying sustainable fashion. Based on reviewed literature and Newman et al.'s (2014) assertion that consumers do not prefer “too green” fashion for fear that quality will be sacrificed, it is hypothesized that:

H2.

Unmet expectations of millennial shoppers positively influence sustainable fashion avoidance intention.

Symbolic incongruence and sustainable fashion avoidance intention

Consumers are known for patronizing brands that enhance their self-concept and avoid those that dilute ideal self-identity perceptions (Sirgy, 1982; Rahman et al., 2020). Symbolic incongruence refers to the inability of a brand to capture the personality facets valued by consumers (Banister and Hogg, 2004). Deindividuation and inauthenticity are identified as two dimensions of symbolic incongruence (Lee et al., 2009). Deindividuation occurs when the brand is not consistent with a consumer's self-concept (Kim et al., 2013). Such a disconnection may result in the brand being consigned to a shopper's inept set (Lee et al., 2009). Millennial shoppers are known to prefer fashion items that enhance their status, personal brand image and self-identity (Vuong and Nguyen, 2018; Iran et al., 2019). For instance, a study by Vuong and Nguyen (2018) found that millennials are more influenced by emotional factors, such as status and symbolism, compared to other cohorts such as baby boomers. As millennials are known to avoid brands they perceive to impair their self-worthiness, it is hypothesized that:

H3.

Symbolic incongruence positively influences millennial shoppers' sustainable fashion avoidance intention.

Materialism and sustainable fashion avoidance intention

Millennial shoppers are described as highly sophisticated and consumption oriented (Eastman and Liu, 2012; Johnstone and Lindh, 2022). From a sustainability perspective, millennials were found to be guided by the dominant social paradigm that is driven by the quest for self-gratification and materialism (Lu et al., 2013; Johnstone and Lindh, 2022). Millennial shoppers are also known to be socialized in a materialistic-oriented society (Bakewell and Mitchell, 2003; Kim and Jang, 2014). It is argued that the consumer trait of materialism has a double effect on millennial shoppers' perception of sustainable fashion. Due to their inclination towards compulsive consumption and accumulation of material possessions (Eastman and Liu, 2012; Dermody et al., 2015), millennials are more likely to prefer fast fashion due to trendiness (Niinimaki, 2010). However, materialistic driven millennial shoppers are also likely to avoid cheap fast fashion as it has the import of diluting their self-esteem (Bhardwaj and Fairhurst, 2010). To convey their materialistic orientation, millennials are more likely to buy sustainable fashion and gain the “green” status associated with paying the premium price associated with sustainable fashion. As millennials were found to exhibit higher levels of materialism (Eastman and Liu, 2012; Butcher et al., 2017), it is hypothesized that:

H4.

Materialism positively influences millennial shoppers' sustainable fashion avoidance intention.

Sustainable fashion avoidance intention and sustainable fashion avoidance

Sustainable fashion avoidance intention measures the extent to which shoppers are willing to avoid buying sustainable fashion (Shen et al., 2014). The performance of environmental behaviours, such as the purchase of sustainable fashion, remains an enigma in many markets. Although millennials were found to report high levels of environmental concern, previous studies revealed that they remain reluctant to buy sustainable fashion (Park and Lin, 2018; Johnstone and Lindh, 2018; Rausch and Kopplin, 2020). In the same vein, the sustainability values of millennial shoppers were found not to be strong enough to trigger purchase behaviour (McNeill and Moore, 2015). According to Johnstone and Lindh (2018), favourable intentions may fail to trigger purchase behaviour due to the unplanned and impulsive buying approach commonly employed by millennial shoppers. Intention to avoid sustainable fashion may also be linked to limited consumer education on the benefits of sustainable fashion (McNeill and Moore, 2015). While intention is regarded as the immediate predictor of behaviour (Ajzen, 1991), it is also known that individuals do not always act based on their stated intentions (Han et al., 2017); thus, it is posited that:

H5.

Sustainable fashion avoidance intention has a positive effect on sustainable fashion avoidance behaviour.

Based on the literature reviewed and hypotheses formulated, Figure 1 presents the conceptual framework.

Methodology

Sample and sampling method

The sample was purposively selected from a cohort of millennial fashion consumers using a web-based online survey. Respondents were recruited from blogs of websites promoting sustainable fashion in South Africa which included SA Fashion Week, Twyg and Fashion Revolution. Millennial shoppers were selected because they are considered to be more environmentally conscious and are more likely to better understand the concept of sustainable fashion than other cohorts (Ruppert-Stroescu et al., 2015). South Africa's Living Standards Measure was used to purposively recruit respondents with the level of income that enables the purchase of the often-premium priced sustainable fashion. Respondents were screened based on their age, knowledge and accessibility to sustainable fashion retail outlets. Millennials were defined as individuals who were born between 1981 and 2000 (Soares et al., 2017; Bento et al., 2018). To be eligible for the study, one had to be gainfully employed and earn a monthly income that fell within the middle-income category. This was done in order to consider only respondents with the financial status that allowed them to pay the premium price associated with sustainable fashion.

Measurement scales and data collection

A structured questionnaire employing a five-point Likert scale was used to collect cross-sectional data. A set of 16 brand avoidance items adapted from Lee et al. (2009), Kim et al. (2013) and Yoon et al. (2020) were used to understand the justifications used by millennial shoppers to avoid sustainable fashion. Materialism was operationalized using a six-item scale adapted from Kim and Jang (2014). A four-item scale adapted from Krishnamurthy and Kuck (2009) was used to measure sustainable fashion avoidance intention. Sustainable fashion avoidance behaviour was measured using a four-item scale adapted from Thompson and Zeynep (2006). Prior to the main survey, the questionnaire was pretested with 60 respondents to assess readability and scale reliability. Data for the main survey was collected over a period of six months from January 2021 to May 2021. A total of 500 questionnaires were distributed and 423 were valid for analysis.

Sample profile

A total of 423 questionnaires completed by millennial shoppers were considered for analysis. The profile of the sample is summarized in Table I.

Preliminary data analysis

The Statistical Package for Social Sciences (SPSS) version 25 software was used for data analysis.

Exploratory factor analysis

In order to determine the dimensions of brand avoidance, exploratory factor analysis (EFA) was conducted. In accordance with the recommendation by Dimirov (2014) to vary the sample size for EFA and that used for model testing, a calibration sample of 160 was used. A sample size of 160 was deemed adequate, based on the Monte Carlo stimulation threshold of N > 150 (Guadagnoli and Velcer, 1988). The suitability of data for EFA was indicated by the Kaiser–Meyer–Olkin (KMO) sampling adequacy (0.803) and Bartlett's sphericity test (p > 0.000). Three items “sustainable fashion is not strong enough”, “sustainable fashion style is dull” and “sustainable fashion uses misleading claims” were dropped due to high cross loadings (> 0.50) and low communalities (< 0.60) (Hair et al., 2013).

A three-factor structure emerged, representing a total variance of 71,622 per cent. Factors were extracted using the Eigen value criterion of greater than 1 (Factor 1 = 8.099; Factor 2 = 5.333 and Factor 3 = 1. 868). Factor 1 was represented by statements related to sustainable fashion values, beliefs and culture and was named ideological incompatibility. Factor 2 was identified as unmet expectations, and it captures the value proposition of sustainable fashion in terms of quality, price and style. Lastly, Factor 3 was identified as symbolic incongruence. The factor comprises statements assessing the relationship between sustainable fashion and brand personality perceptions of respondents. Table II provides the factor structure of brand avoidance dimensions and the item loadings.

Data normality and common method bias

Using a total sample of 423, data normality was assessed using the Kolmogorov–Smirnov (K–S) and Shapiro–Wilk (S–W) tests. The K–S and S–W test values were all insignificant (p > 0.05), indicating the normality of data (Hair et al., 2013). Self-reported data was collected from millennial shoppers which made the assessment of the potential influence of common method bias a necessity. Following Harman's single factor procedure, all scale items that were used to measure the study constructs were subjected to an un-rotated factor analysis (Gaskin, 2011). This procedure resulted in a single factor with a variance of 44.2 per cent which is below the recommended cut-off point of 50 per cent, an indication that common method bias was not a problem in this study (Podsakoff et al., 2012).

Reliability assessment, correlational analysis and hypotheses testing

Measurement scale reliability and validity

Cronbach's alpha coefficients were computed to assess reliability and validity of the scale items used for the main survey. As shown in Table III, all coefficients were above the acceptable cut-off point of 0.7, confirming the reliability of the measurement scales (Hair et al., 2013). Moreover, the item-total correlations were all above 0.5, indicating the attainment of convergent validity (Pallant, 2011). Table III summarizes the descriptive statistics, reliability and validity indicators.

Correlation analysis

The Pearson product-moment was used to examine the degree of association, strength and direction of relationships between study variables. The correlation coefficient (r) values were all positive and significant, ranging from small to strong relationships. The correlation coefficients were all below 0.8, indicating the absence of multi-collinearity and attainment of discriminant validity (Pallant, 2011; Hair et al., 2013). Table IV provides the correlation matrix.

Hypotheses testing results

Standard multiple regression analysis was used to test proposed hypotheses. This statistical technique was preferred because it permits the estimation of multiple linear relationships while controlling correlation residuals (Pallant, 2011). The sample size of 423 was adequate for conducting regression analysis based on Tabachnik and Fidell's (2007) criteria N > 50 + 8m, where m = number of independent variables. Multi-collinearity and singularity were not an issue as correlation coefficients were all below the recommended threshold of +0.80 and −0.80 (Grewal et al., 2004). Tolerance values and variance inflation factor values were above 0.1 and less than 10, indicating non-collinearity (Pallant, 2011). Model 1 included ideological incompatibility, unmet expectations, symbolic incongruence and materialism as independent variables and sustainable fashion avoidance intention as the dependent variable. The model yielded an R-square value of 0.502, indicating that independent variables explained approximately 50 per cent of the variance in sustainable fashion avoidance intention.

The Durbin Watson test statistic was computed to assess the possibility of autocorrelation. Autocorrelation occurs when independent variables have a causal influence on the dependent variable within the regression model (Arjmand and Shafiei, 2018; Turner, 2020). The values of the Durbin Watson test range from 0 to 4, with values between 1.5 and 2.5 signifying the absence of autocorrelation (Chapman et al., 2004; Arjmand and Shafiei, 2018). As shown in Table V, the Durbin Watson statistic for Model 1 was 2.013, indicating the absence of autocorrelation. Table V summarizes the results of regression analysis.

As shown in Table V, H2, H3 and H4 were supported, while H1 was not supported. Based on path coefficients, unmet expectations emerged as the major predictor of sustainable fashion avoidance intention, followed by symbolic incongruence while the effect of ideological incompatibility was insignificant.

Model 2 tested the influence of sustainable fashion avoidance intention on sustainable fashion avoidance. The model showed an R-square value of 0.262, indicating that sustainable fashion avoidance intention accounted for 26.2 per cent variance in sustainable fashion avoidance behaviour. As shown in Table VI, the Durbin Watson statistic was 1,833, indicating that autocorrelation was not a problem (Arjmand and Shafiei, 2018). Model 2 results are shown in Table VI.

Discussion of results

The first hypothesis (H1) which predicted a positive relationship between ideological incompatibility and sustainable fashion avoidance intention (H1) was not supported (β = 0.019, t = 1.164, p < 0.061). The plausible explanation for this result could be that the concept of sustainable fashion in South Africa is still evolving, and related ideologies may still be in the formative stages. This finding affirms the plurality of views related to sustainable fashion (McNeill and Moore, 2015; Vuong and Nguyen, 2018). This finding points to the need to develop value propositions that align sustainable fashion with the cultural values and beliefs valued by millennial shoppers.

The second hypothesis (H2) which posits a positive effect of unmet expectation on sustainable fashion avoidance intention was supported (β = 0.345, t = 8.839, p < 0.001). This result suggests that South African fashion consumers perceive that sustainable fashion is unable to meet their expectations. This finding concurs with previous studies in which unmet expectations were found to result in brand avoidance (Morgan and Birtwistle, 2009; Newman et al., 2014; Armstrong et al., 2015). For instance, in a study conducted by Armstrong et al. (2015), young Finnish consumers avoided sustainable fashion due to lack of attributes, such as novelty and trendiness. In a study conducted by Morgan and Birtwistle (2009), lack of education on the concept of sustainability was found to foster perceptions of unmet expectations. A study by Newman et al. (2014) also found that fashion consumers are more likely to develop intentions to avoid sustainable fashion if they perceive that they are forced to sacrifice core benefits, such as affordability and quality. Consumer education on the benefits of sustainable fashion is recommended as a long-term behavioural change strategy. Designers and retailers of sustainable fashion may respond to the need for trendiness by decreasing the lead times, being more responsive to consumer demands and broadening product variety.

The third hypothesis (H3), which postulated a positive effect of symbolic incongruence on sustainable fashion avoidance intention, was supported (β = 0.274, t = 5.228, p < 0.001). This result suggests that South African fashion consumers perceive sustainable fashion to offer little symbolic meaning. This result is consistent with findings from previous studies (Ciasullo et al., 2017; Kozlowski et al., 2019). For instance, Ciasullo et al. (2017) found that consumers give little significance to the concept of sustainability when making fashion purchase intentions. In the same vein, Kozlowski et al. (2019) note that the limited focus on personalization and cultural dimensions by sustainable fashion designers reduce the market appeal of environmentally friendly apparel. Marketers of sustainable fashion in South Africa may enhance symbolic congruence by incorporating brand personality facets valued by millennial consumers, such as novelty, trendiness and self-expressiveness (Park and Lin, 2018; Jung et al., 2016). According to Chakraborty and Sadachar (2022), this can be achieved by developing marketing messages that incorporate cultural cues that evoke favourable attitudes towards sustainable fashion.

The fourth hypothesis (H4) examined the effect of materialism on sustainable fashion avoidance intention (β = 0.332, t = 8.215, p < 0.001). This result suggests that consumers with a materialism orientation are less likely to buy sustainable fashion. This result is consistent with findings from previous studies that found that personality of traits of millennials were found to influence eco-fashion purchase behaviour (Fu and Liang, 2018; Heo and Muralidharanm, 2019). For consumers with a materialistic disposition, fast fashion, the alternative of sustainable fashion, serves as a visible tool to portray their wealth and status (Yoon et al., 2020). This result suggests the need by South African marketers of sustainable fashion to utilize the “green” status associated with the consumption of sustainable fashion as a key selling proposition to target millennial shoppers.

The fifth hypothesis (H5) predicted that sustainable fashion avoidance intention positively influences sustainable fashion avoidance behaviour. The hypothesized relationship was confirmed (β = 0.193, t = 1.041, p < 0.05). This finding is consistent with the theory of planned behaviour which posits a positive relationship between intention and behaviour (Ajzen, 1991). It should be noted that this relationship is weak as indicated by a weak significance value. This result suggests that fashion consumers do not hold predetermined intentions or deliberately plan to avoid buying sustainable fashion. Thus, as suggested by Johnstone and Lindh (2018), marketers of sustainable fashion may need to utilize situational cues such as affordable prices and awareness campaigns. Sustainable fashion awareness campaigns are important as South African shoppers experience challenges in identifying fashion incorporating environmental labels (Dreyer et al., 2016).

Conclusions, managerial implications and limitations

A modified BAM was applied to examine factors that influence sustainable fashion avoidance behaviour among millennial South African shoppers. Unmet expectations, materialism and symbolic incongruence emerged as major predictors of consumers' intention to avoid sustainable fashion. In order to satisfy fashion consumers' expectations, marketing managers need to enhance the quality of sustainable fashion. The study findings point to the need to align sustainable fashion with brand personality elements valued by millennials. The need to enhance brand personality is based on Rath and Bay's (2015) observation that consumers prefer apparel aligned to their personal style and values. Consumers' values may be incorporated by considering the cultural architecture valued by consumers. This will go a long way in personalizing and customizing sustainable fashion, elements which are perceived to be lacking in sustainable fashion offerings (Kozlowski et al., 2019). Sustainable fashion marketers may also consider investing in educational campaigns to reinforce positive attitudes towards sustainable fashion. Such educational campaigns, as suggested by Morgan and Birtwistle (2009), may focus on the value proposition of sustainable fashion while also emphasizing the environmental harm caused by fast fashion consumption. In this regard, designers are urged to develop eco-labels that succinctly communicate the environmental benefits of sustainable fashion.

Although this study provides valuable insights, it is not immune to limitations. The study relied on cross-sectional data. This limited the study in tracking the possibility of changes in attitudes towards sustainable fashion in the long term. Future studies may employ a longitudinal time horizon in order to fully understand the concept of sustainable fashion as it evolves. The study utilized a quantitative research method and relied on close-ended measurement scales. This study may be extended in future studies by employing a mixed-method approach. The qualitative data has the potential to produce deeper insights related to the concept of sustainable fashion. The study also focused only on millennial shoppers. Future studies may also consider other cohorts to ensure comparability of results. Lastly, the study was confined to one emerging market – South Africa. Future studies may also focus on other emerging markets, such as India, Brazil and China. Such a cross-market study has the potential to accord marketers of sustainable fashion valuable insights to position sustainable fashion in international markets.

Figures

Figure 1.

Conceptual framework

Table I.

Sample profile

Gender
Male	47% (n = 199)
Female	53% (n = 224)
Ethnicity
Black African	29% (n = 123)
Indian/Asian	43% (n = 180)
White	28% (n = 120)
Education level
Matriculation certificate	8% (n = 34)
Diploma	21% (n = 89)
Degree	49% (n = 207)
Masters	15% (n = 63)
Doctoral degree	7% (n = 30)
Age
25–30 years	23% (n = 97)
31–35 years	32% (n = 135)
36–41 years	45% (n = 190)
Monthly income
R40,000–R50,000	17% (n = 72)
R51,000–60,000	32% (n = 135)
Above R61,000	51% (n = 216)

Table II.

Factor structure of sustainable fashion brand avoidance dimensions

Scale items	Brand avoidance dimensions: item loadings
	Factor 1	Factor 2	Factor 3
Sustainable fashion makes the world's fashion look the same	0.821	0.221	0.011
Sustainable fashion is not consistent with my culture	0.858	0.140	0.063
Sustainable fashion ruins my fashion beliefs	0.901	0.236	−0.112
Sustainable fashion is against my values	0.888	0.273	−0.054
Quality of sustainable fashion is not good enough	0.078	0.864	0.233
Price of sustainable fashion is too high	0.023	0.826	0.112
Sustainable fashion lacks style	−0.054	0.912	0.013
Sustainable fashion lacks uniqueness	0.052	0.902	0.064
Sustainable fashion uses cheap recycled material	0.011	0.890	0.022
Sustainable fashion makes it hard for me to express my personality	0.017	0.194	0.784
Sustainable fashion does not enhance my self-identity	0.29	0.067	0.761
Sustainable fashion styles have too much mass appeal	0.32	0.220	0.782
Sustainable fashion does not support my fashion style	0.024	0.217	0.822
Bartlett's Test of Sphericity		df	256
		Sig.	0.000
Kaiser–Meyer–Olkin measure of sampling adequacy			0.803

Notes: Factor 1 = ideological incompatibility; Factor 2 = unmet expectations and Factor 3 = symbolic incongruence. Sig.: significance

Major loadings for each item are given in italics

Table III.

Descriptive, reliability and validity statistics

Construct	Scale items	Source	Mean	Item-total correlation	Cronbach's alpha
Ideological incompatibility	Sustainable fashion makes the world's fashion look the same	Kim et al., 2013; Yoon et al., 2020	3.852	0.726	0.860
	Sustainable fashion is not consistent with my culture		3.984	0.715
	Sustainable fashion ruins my fashion beliefs		3.865	0.637
	Sustainable fashion is against my values		3.831	0.749
Unmet expectations	Quality of sustainable fashion is not good enough	Kim et al., 2013; Colucci and Scarpi, 2013	4.04	0.673	0.914
	Price of sustainable fashion is too high		4.34	0.582
	Sustainable fashion lacks in style		3.98	0.568
	Sustainable fashion lacks in uniqueness		4.02	0.684
			4.52	0.670
Symbolic incongruence	Sustainable fashion makes it hard for me to express my personality	Lee et al., 2009; Kim et al., 2013.	3.594	0.576	0.731
	Sustainable fashion does not enhance my self-identity		3.882	0.685
	Sustainable fashion styles are dull		3.650	0.770
	Sustainable fashion does not support my fashion style		3.850	0.667
Materialism	I admire people who buy trendy fashion	Kim and Jang, 2014; Lang and Armstrong, 2018	4.22	0.711	0.933
	The way I dress say a lot about how I am doing in life		3.88	0.699
	I like to buy fashion that impress my peers		4.35	0.712
	It gives me pleasure to try various fashion styles		4.22	0.745
	I value possessing latest fashion		4.28	0.750
Sustainable fashion avoidance intention	I intent to avoid buying sustainable fashion	Krishnamurthy and Kucuk, 2009	3.752	0.588	0.836
	I plan to avoid buying sustainable fashion in future		3.696	0.680
	I intent to expend more effort looking for fast fashion		3.761	0.707
	I intent to buy fast fashion each time I shop		3.826	0.695
Sustainable fashion avoidance behaviour	I buy fast fashion every time I go for shopping	Thompson and Zeynep, 2006	3.088	0.757	0.838
	I avoid buying fashion labelled as sustainable		3.090	0.654
	I refuse to buy sustainable, instead I opt fast fashion		3.411	0.652
	I always prefer to buy fashion		3.139	0.648

Table IV.

Correlation matrix

Construct	ID	SI	UE	SFAI	MAT	SFA
ID	1.00
SI	0.574*	1.00
UE	0.652**	0.562**	1.00
SFAI	0.142*	0.547**	0.645**	1.00
MAT	−0.113*	−0.233*	−0.122*	−0.453**	1.00
SFAB	0.124*	0.317**	0.302**	0.339**	−0.572**	1.00

**

Notes: p < 0.001,

*

p < 0.05

ID: ideological incompatibility, SI: symbolic incongruence, UE: unmet expectation, SFAI: sustainable fashion avoidance intention, MAT: materialism, SFA: sustainable fashion avoidance behaviour

Table V.

Regression model 1

Independent variables: ID, UE, SI and MAT	Dependent variable: SFAI
	Unstandardized coefficients		Standardized coefficient	Sig.		Collinearity statistics		Durbin Watson value
	Β	Std. error	Beta	T	Sig.	Tolerance	VIF
Constants	30.272	0.983		31.603	***			2.013
ID	0.094	0.114	0.019	1.164	0.061	0.923	1.067
UE	0.643	0.059	0.345	8.839	***	0.812	1.210
SI	0.361	0.033	0.274	5.228	***	0.651	1.480
MAT	0.622	0.061	0.332	8.217	***	0.719	1.378

Notes: R² = 0.502; adjusted R² = 0.491

ID: ideological incompatibility, SI: symbolic incongruence, UE: unmet expectation, MAT: materialism, SFAI: sustainable fashion avoidance intention, Sig.: significance, Std.: standard. ***p < 0.001

Table VI.

Regression model 2

Independent variable: SFAI	Dependent variable: SFAB
	Unstandardized coefficients		Standardized coefficient	Sig.		Collinearity statistics		Durbin Watson value
	Β	Std. error	Beta	T	Sig.	Tolerance	VIF
Constants	2.158	0.153		9.153	***			1.833
ATSF	0.245	0.038	0.193	1.041	*	0.880	1.00

Notes: R² = 0.262; adjusted R² = 0.256, ***p < 0.001, *p < 0.05.

SFAI: sustainable fashion avoidance intention, SFAB: sustainable fashion avoidance behaviour, Sig.: significance, Std.: standard