RESEARCH ARTICLES

 

Exploring the factors that impact attitude to purchase: A spotlight on counterfeit luxury handbags among Durban's emergent black middle-class females

 

 

Mishka Jugnundan^I; Pragasen Pillay^II

^ISchool of Management Studies, University of Cape Town, South Africa Email: mishka.iuqnundan@gmail.com; ORCID: https://orcid.org/0000-0002-9331-197X
^IISchool of Management Studies, University of Cape Town, South Africa. Email: p.pillay@uct.ac.za; ORCID: https://orcid.org/0000-0001-8511-6990

 

 

---

ABSTRACT

PURPOSE OF THE STUDY: This study investigates the factors that impact attitude to purchase, with a product focus on counterfeit luxury handbags among Durban's emergent Black middle-class females
DESIGN/METHODOLOGY/APPROACH: The investigation followed a quantitative approach whereby 350 individuals provided data through self-administered questionnaires. Data was interpreted through factor and regression analysis
FINDINGS: The research findings confirmed that individual product and service (IPS) factors had an impact on the attitude to purchase. Furthermore, it was discovered that knowledge, perceived risk, ethical obligation, price, and service quality significantly influenced attitude formation and, consequently, attitude to purchase
RECOMMENDATIONS/VALUE: From the demand perspective, i.e., via the IPS factors, the findings of the study assist anti-counterfeit bodies in tailoring their efforts to curb counterfeit purchases
MANAGERIAL IMPLICATIONS: Policymakers and managers alike are required to be more aware of the impact of demand-related factors when formulating effective strategies to deal with the issue of counterfeit purchases
JEL CLASSIFICATION: M30, M31

Keywords: Attitudes; Counterfeit luxury handbags; Emergent black middle class, South Africa.

---

 

 

1. INTRODUCTION

Counterfeiting is a significant transnational concern occurring in both developed and emerging countries. Although efforts have been channelled towards discovering the source of counterfeit goods produced by various governmental and legal entities, little emphasis has been placed on the buyers of these products or the motivation behind the perceived purchase behaviour (Stöttinger & Penz, 2017). In recent years, the city of Durban, KwaZulu-Natal, has experienced a rapid influx of counterfeit cargo seizures, many of which have included fake luxury-branded handbags (Nair, 2012; Nxumalo & Mthethewa, 2016). Therefore, the sheer volume of this ongoing supply illustrates the demand for these goods, which prompts another question: Who is purchasing these products? This, in turn, raises questions surrounding who counterfeit consumers are.

In terms of its population, the majority of the citizens of Durban are black and female. Furthermore, the city is home to a large EBMC Emergent Black Middle-class group (EBMC) (StatsSA, 2017). The conspicuous consumption habits of the group, in combination with monetary restrictions and established middle-class aspirations, make these individuals an interesting target market from which to extract data in relation to counterfeit luxury-branded handbags (Burger et al., 2014). By obtaining a greater understanding of the factors that impact the attitude to purchase of these consumers, more informed efforts to combat counterfeit purchases can be constructed, and entities, such as SARS (South African Revenue Services), will benefit as a result. Prior to the commencement of this research investigation, the relationship between EBMC consumers' attitudes towards 'fake' high-end handbags and purchase behaviour remained unknown.

 

2. REVIEW OF THE LITERATURE

The literature review first examines the theory on which the study's conceptual framework was based. Once an understanding of the framework is established, the contextual setting of the research investigation is presented, more specifically, an overview discussing counterfeit luxury goods and the EBMC.

2.1 The theoretical framework

The framework utilised for the study draws from several established attitudinal models. The concept of attitude is therefore primarily discussed, followed by of review of significant existing theoretical models. Lastly, the conceptual framework is presented, which outlines the links between the variables of the study and depicts the hypotheses of the investigation.

2.1.1 Attitude

Attitude has historically occupied a central role within the realm of consumer behaviour studies (Hanssens et al., 2014). As such, attitude has been utilised within varying research investigations pertaining to different elements of the concept, ranging from attitudinal formation to changes in attitude and the impact of attitude on purchase decisions (Babin & Harris, 2015; Meyer, 2019). Thus, multiple definitions of attitude have surfaced, with theorists being unable to agree upon a single, standardised meaning of the concept. According to Baggozzi et al. (2002), however, the most extensively welcomed description of attitude considers it to be an evaluation. Hung et al. (2016) concur with Baggozzi et al. (2002), citing that attitude may be described as a positive or negative reaction trend towards specific products, which may be developed over time from previous experiences. The theorists also infer that relationships may be evident between consumer attitude and future purchasing behaviour (Hung et al., (2016). The definition of attitude presented below, proclaimed by Hung et al. (2016), alludes to the nature of the study in that it insinuates the relationship evident between attitude and purchase behaviour. "Attitude is described as a positive or negative reaction trend towards specific products, which may be developed over time from previous experiences and have an influence on future behaviour." In order to explain the underlying dynamics evident between attitude and buying behaviour, various structural models have been proposed. These models typically segment the concept of attitude into attributes, which aid in providing more focused explanations or predictions of consumer behaviour. Several investigations have been conducted in order to better comprehend the dynamics underlying attitude, attitude formation, as well the relationship between consumer attitude and buyer behaviour. Thus, theorists have proposed various structural models of attitude to identify attitudinal dimensions and to explain or predict consumer behaviour.

Among these models are the tri-component model, as well as various multi-attribute structures (all of which assume a rational model of human behaviour) (Schiffman et al., 2013). The tricomponent model broadly deals with three fundamental attitudinal components, namely; (cognitive, affective and conative). According to Albarracin and Johnson (2018), the cognitive component refers to the consumer's knowledge and perceptions generated through direct experience with an object or through their interactions with others, while the affective aspect of attitude consists of the consumer's emotions or feelings about a product or service. The final component of this model, the conative element, relates to the likelihood or tendency that a consumer will act in a specific manner regarding an object, that is, their intent to purchase (Albarracin & Johnson, 2018).

Multi-attribute attitude models, in comparison, tend to examine the composition of consumer attitudes with regard to specific products and beliefs (Babin & Harris, 2015). Models that fall within this category include the Theory of Reasoned Action (TRA) model, the Attitude-Toward-Object (ATO) model, and the Attitude-Toward-Behaviour (ATB) model, amongst others (Schiffman & Kanuk, 2004). In light of the fact that this study primarily dealt with a specific product (counterfeit luxury-branded handbags), elements derived from the Attitude-Toward-Object (ATO) model as well as the Theory of Reasoned Action (TRA) model, were included. These elements include 'ethical obligation' and 'perceived risk' associated with the TRA, along with 'product price' and 'product quality' aligned with the ATO model.

Although the Ethical Decision Making (EDM) theory model is not widely used when undertaking attitude-based research, the context of the study added a moral dimension on account of the illegal nature of the products concerned, whereby consumers may have encountered the aspect of an added ethical paradigm in contrast to their regular purchase consideration process.

The EDM model demonstrates that multifaceted variables affect the likelihood of ethical actions by individual decision-makers (Schwartz, 2016). The framework illustrates the coupling of individual and organisational factors as a means of influencing individuals faced with an EDM challenge (Shapiro & Stefkovich, 2016). Though both organisational and individual factors are regarded as important in terms of their contribution towards decision-making, several authors contend that it is individual factors that play a greater role in terms of the decision-making process (Frederickson & Ghere, 2013). Wittmer (2016), for example, posits that the cognitive moral development of an individual is a critical variable in terms of explaining EDM behaviour. As such, the individual factor of 'knowledge' included in the EDM model took precedence over the elements of the ATB model as the study dealt primarily with individuals making a purchase decision, as opposed to forming an evaluation regarding proceeding with one. The EDM theory model was therefore incorporated within the study's theoretical framework through the 'knowledge' variable.

The inclusion of the EDM model, therefore, took precedence over the ATB model due to its innate connection with the context of the study.

The conceptual framework crafted for the purposes of addressing the research question at hand draws both on established theoretical foundations as well as insights generated from more current investigations (Malhotra, 2017). In addition to making use of elements stemming from well-established models, the researcher chose to include 'service quality' as an additional intervening variable given the results of a local study conducted by research firm Nielsen. Service quality was integrated due to evidence of its significance in relation to South African purchase behaviour in particular. Evidence from the report indicated that South African customers wanted not only to search for and receive products quickly and conveniently but also to find answers to their questions or problems easily, and if not, they were quick to move away from a business or specific brand (Nielsen, 2015). Thus, within the South African context, the quality of the experience one receives may be considered to be integral to the purchase decision-making process. In order to test this premise, 'service quality' was included as a variable in the study. While each of the framework's six intervening variables may be viewed as distinctive in its own right, they may be grouped to form three categories, namely, individual factors, product factors and service factors. Ultimately, it is the presence or lack of these factors that result in the attitude to purchase (regarded as the output variable) (Brennan et al., 2014). The conceptual framework utilised for the purpose of this study illustrates the aforementioned variables and is presented hereafter in Figure 1.

2.2 The contextual setting: counterfeit luxury goods and the emergent black middle class

Historically, counterfeit goods were considered to be easily identifiable as they generally represented 'luxury' products manufactured from substandard materials, often traded in a limited number of cosmopolitan city locations (Waring, 2013). Today, counterfeiting affects virtually every product category, ranging from the production of imitation food and beverage items to pharmaceuticals, consumer electronics and even automobile parts (IACC, 2017). As a result, the definition of counterfeits has broadened over time to encompass all possible product replication variations. According to Chiu and Leng (2016), counterfeit branded goods refer to an illegally manufactured copy of a genuine brand, whereas imitation relates to the legal manufacturing of look-alikes (including generics) (OECD, 2016). In light of the above descriptions, the current investigation pertained solely to counterfeit branded goods, more specifically, counterfeit luxury-branded handbags. In recent years, the demand for accessories (including items such as belts, wallets, pens and handbags) has rapidly increased due to their fashion versatility (Freer, 2018). Although genuine-brand accessory sales have been boosted because of such demand, not all consumers are able to satisfy their desires, with income serving as one of the key inhibiting factors (Atwal & Bryson, 2017).

Choi and Shen (2016), therefore, contend that the demand for branded accessories (more so those with distinct features, such as those designed by luxury brand houses) continues to exist on a global scale. The existence of this demand, in combination with the influence of various socio-economic factors, has thus allowed counterfeiters to prosper through the sale and manufacture of goods that meet both the preferences and budgetary requirements of the bulk of consumers (Fabrizio, 2016). In 2012, for example, 500 million handbags, belts and wallets with a street value of $1 billion were confiscated in the United States of America (USA) alone (Pohlman & Day, 2013).

Though a significant amount of counterfeit accessory purchases takes place in 'first world' nations, Som and Blanckaert (2015) maintain that the demand for these products is also widespread and evident in developing countries. In fact, Rathore (2013) reported that the Indian luxury replica product market is increasing by up to 40 percent per annum, with demand for counterfeit shoes, apparel and handbags having even exceeded the demand for genuine designer items on the subcontinent. Furthermore, in 2011, government reports indicated that the counterfeit goods market in Turkey doubled to USD 6 billion, with the industry having been valued at half this amount (roughly USD 3 billion) in the previous year (Letsch, 2011). Of these goods, fake luxury-branded handbags were named the most widely manufactured and sold product in the country (Beckert & Musselin, 2013). Though counterfeit products are frequently sold to consumers within the borders of the country of their origin, Adelman et al. (2014) argue that many customers often also purchase these goods whilst abroad, with the presence of counterfeit markets having even been cited as reasons for tourist travel in recent years. The findings of a recent British consumer report concur with the opinion of the above-mentioned authors, reporting that almost half of the travellers surveyed purchased counterfeit items overseas, with the majority of consumption activity having taken place in Turkey, Greece, Spain, Thailand and China (Bell, 2016). Moreover, the report stipulated that the most popular item among the counterfeit goods purchased was luxury-branded handbag replicas (Bell, 2016).

Evidence from India and Turkey, for example, has recorded an increase in replica product markets by 40% or more, fuelling billions in counterfeit turnover globally (Letsch, 2011; Rathore, 2013). Of the illegal goods sold, counterfeit luxury handbags are regarded to be the most popular in both manufacturing and sales quantities (Beckert & Musselin, 2013; Bell, 2016). Interestingly, evidence stemming from research conducted on counterfeit activity within Asian markets has depicted a similar landscape regarding the demand for certain goods (Paquette, 2018). Counterfeit Louis Vuitton handbags, in particular, have been named as the most widely seized goods in South Korea for several years (Pecotich & Shultz, 2016). Comparably, in Japan, counterfeit luxury-branded handbags accounted for the majority (55%) of all customs confiscations in 2015 (Thomas, 2015).

Although the most commonly counterfeited brands tend to vary slightly from nation to nation, Sun et al. (2015) document that across the board, higher-end designers account for most mass-produced replica bags. According to the hierarchy of luxury brands, which was developed by Megan Willet in 2015 (Willet, 2015), these brands typically fall into the 'accessible' to 'premium' core of the market and include names such as Louis Vuitton, Prada, Chanel, Burberry, Hermes and Gucci.

Despite the generally negative sentiment surrounding the sale of counterfeit luxury-branded handbags by law enforcement officials and the like, some authors contend that the distribution of these products does, in fact, have some positive repercussions for genuine brands (Kapferer, 2015). Raustiala and Springman (2012), for example, maintain that creativity thrives in the face of counterfeit activity, with design imitations forcing brands to innovate more quickly. In his research, Amaral (2016) also argues that the sale of counterfeit goods increases levels of brand awareness of original brands among consumer groups. On the contrary, however, luxury brand houses have vigorously defended their view on counterfeit production, with some brands even citing the counterfeit industry as an 'ongoing burden', weighing heavily on their efforts to deliver on promises of exclusive offerings (Shams, 2015).

As a result, some luxury brands have taken it upon themselves to take direct interventions to defend their intellectual property rights in their own capacities (Yang, 2015). Louis Vuitton, ranked as the world's most valuable luxury brand in 2015, is considered to have taken the most aggressive stance where the battle against counterfeits is concerned, initiating more than 10 000 raids annually worldwide (Thomas, 2015; Bell, 2016).

Yet, despite the concerted efforts of genuine brands such as Louis Vuitton, it remains relatively easy for consumers to purchase counterfeit handbags (Stewart, 2014). Moreover, the availability and ease of access to supply have been exacerbated by technology (such as internet selling platforms), making counterfeit merchants harder to track (McNeil & Riello, 2016). The occurrence of this phenomenon, in combination with economic growth in developing nations reaching new heights, has propelled the sale of counterfeit goods in new markets, with Africa being no exception to the infiltration (Lee, 2014).

South Africa's middle class is far from being as wealthy as its 'top-end'; however, this segment of the population has experienced significant growth since 1994, more so among black citizens (Alexander et al., 2013). In fact, while the white middle class has shown signs of regression (shrinking from 2.8 million adults in 2004 to 2.68 million adults in 2015), the black segment grew by 3.7 million (from 1.6 million to 5.3 million) adults over the same time period, indicating an astounding growth rate of 280 percent (Southhall, 2016).

Aside from an interest in the black middle class stemming from population growth, researchers, marketers, and businesses alike have also expressed interest in this group because of its unique characteristics (Khunou, 2017). Many members of the black middle class are known to possess an asset deficit, thus being continually required to play 'asset catch-up' with regard to their spending. With individuals falling in this group often being the first members of their families in an elevated financial position, members of this segment thus make substantive purchases, such as cars or property, for the first time (Simpson & Lappeman, 2017). Though these purchases somewhat diminish the amount of disposable income available to individuals, their remaining income is often spent on leisure activities, including shopping, takeaway meals and social events (Iqani, 2015). Kotze et al. (2016) cites the overall spending power of the black middle class to be valued at approximately R440 billion, in contrast to that of the white middle class, which is estimated to be R360 billion.

Aside from their consumption behaviour, the black middle class is known to value education deeply, viewing it as a fundamental tool towards bettering one's future (Simpson & Lappeman, 2017). Almost 50 percent of members belonging to the black middle class possess a post-matric qualification, with the number of black middle-class tertiary graduations increasing by over 1.5 million since 2004 (Kotze et al., 2016). In addition, members of the segment have been deemed to value convenience and connectivity. With many of these consumers leading fast-paced lifestyles, they are often willing to move towards product and service offerings which save them time (Iqani, 2015). This sense of convenience is further supplemented through means of digital connectivity, with almost all adults in this segment possessing a mobile device (Simpson & Lappeman, 2017). It is, therefore, evident that, overall, the black middle-class forms part of an interesting segment within the South African population because of the unique social standing, volume and economic positioning of the group (Southhall, 2016). Yet, due to the sheer magnitude of the segment, large disparities continue to exist among the group. Khunou (2017) thus argues that all individuals who represent part of the black middle class cannot be viewed in a homogenous manner. Multiple local researchers agree with the sentiments expressed by Khunou (2017) and have therefore sought to further divide the group into segments that display similar characteristics.

When assessing the black middle class in its entirety, it is evident that two major differing segments emerge. Researchers have coined these segments as the 'established' and 'emerging' members of the middle class (Kotze et al., 2016). While both 'established' and 'emerging' consumers provide a basis for interesting research, some socio-economic characteristics predispose specific middle-class individuals towards making certain purchase decisions (Ncube & Lufumpa, 2015).

Thus, each of these black middle-class segments may display differing purchase behaviour when confronted with the opportunity to obtain counterfeit luxury-branded handbags. In this case, however, emergent consumers may be more likely to purchase the product under consideration compared to their established peers (Burger et al., 2014). Mattes (2014) states that in order to begin to understand the emergent segment of the South African black middle class (EBMC), it is necessary first to gain insight into those deemed to be their 'more successful' counterparts, that is, the established South African black middle class. Research conducted by Burger et al. (2014) indicates that both the established black and white middle classes display similar productive characteristics. It is estimated that half of the heads of households that fall within these two clusters have, at minimum, a diploma, while more than 20 percent have a degree qualification. Moreover, the heads of these households are typically older and more well-established regarding their occupations, mirrored by their high asset index scores (Southhall, 2016). The most noticeable difference between the two groups, however, is the much higher income level (per capita) of white household heads, which has been attributed to age, historical representation in financially rewarding positions and a greater reliance on passive income streams, among other factors (Simpson & Lappeman, 2017).

Aside from differences apparent between the established black and white middle classes in terms of income, Southhall (2016) notes that monetary disparities also exist between the established and emergent black middle classes. The established black middle class (which comprises about 2.57 percent of the total South African population) displays an average income per capita of R120 603, while the emergent group constitutes roughly 2.97 percent, and exhibits an average income per capita of R97 036 (Burger et al., 2014). In addition to being less affluent than the established black middle class, heads of emergent black middle-class households are deemed to be younger, less experienced and less likely to possess a tertiary educational qualification than their peers (Southhall, 2016). Furthermore, EBMC households are more likely to be headed by single women (Khunou, 2017). Thus, the asset index scores of those falling within the EBMC segment are regarded to be much lower than that of both the established black and white middle classes (Burger et al., 2014).

Southhall (2016) contends that although some individuals that form part of the emergent group may transition to the established segment over time (through means such as education and occupation promotions), a significant portion of the group is structurally less advantaged and is more likely to remain within the emergent group. Burger et al. (2014) cite the lack of access to opportunities within urban labour markets and the absence of spousal income to be key factors contributing towards this occurrence.

Despite such obstacles, however, the emergent segment of the black middle class remains one of the most targeted South African markets because of both their buying power and unique consumption habits (Simpson & Lappeman, 2017). Interestingly, evidence has shown that the emergent black middle-class group allocates a greater share of spending to conspicuous consumption items (10.67 %) than the established black (7.79 %) and white middle classes (4.99 %) (Simpson & Lappeman, 2017). Furthermore, a report compiled by researchers at the University of Stellenbosch found that members of the emergent black group were more likely to make use of visual cues to distinguish themselves from their reference groups, spending a considerably larger portion of their income on consuming products that noticeably signal affluence (Burger et al., 2014). It is, therefore, the combined distinctive socio-economic characteristics and unique positioning of the EBMC which define the segment as an attractive market for legal business operators and counterfeit merchants alike.

Although members of the EBMC are dispersed across the nation, individuals located within port cities, such as Durban, may display a greater sense of exposure to counterfeit products and, in turn, possess more favourable attitudes towards counterfeit luxury-branded handbags and the associated purchasing thereof. According to the last national census conducted by StatsSA (2011), Durban's total population currently stands at 595 061, the majority of which are black and female. Although there is no direct recollection of evidence in terms of the buyers of counterfeit luxury-branded handbags within the Durban area, the sheer magnitude of the supply of these goods indicates the existence of substantial demand among the population. Generally, the nature of the goods concerned (handbags) tends to be more aligned with the female gender, and they are marketed for female consumption (Tomshinsky, 2016). The EBMC female group may therefore be inherently regarded as a key market for counterfeit sellers because of their gender. However, additional key characteristics of the group further predispose the segment towards making counterfeit-linked purchase decisions (Iqani, 2015). Their income levels and innate desire to display evidence of conspicuous consumption, for example, too, pre-empt counterfeit consumer behaviour (Burger et al., 2014).

In other words, the demographics of the group, in combination with their socio-economic standing align with that of 'would-be' luxury-branded counterfeit handbag consumers (Kotze et al., 2016). Thus, given the way in which their unique positioning combining socio-economic aspirations and limited financial flexibility, create a viable avenue for luxury-branded handbag counterfeit purchases, emergent black middle-class women may be considered to be both a suitable and relevant sample in relation to the nature of the investigation. In order to truly examine the legitimacy of this claim and to determine the attitude of EBMC female consumers towards counterfeit handbags, however, a research investigation was required to be initiated (Malhotra et al., 2017).

 

3. RESEARCH METHODOLOGY

The following section will delve deeper into the methodology utilised for the investigation and discuss elements such as the research paradigm and design, ethical considerations, the target population and sampling approach and lastly, the research instrument, hypothesis formulation and data analysis.

3.1 Research paradigm and design

While positivists deem scientific measures as a means to both classify and understand phenomena, post-positivist approaches suggest that though the positivist school of thought may be true to a certain extent, interpretations, particularly those made within a societal context, may be based on assumptions and conjecture (Howell, 2012; Barkway, 2013). Post-positivists, therefore, posit that while the physical world may be stable, living beings exhibit an element of uncertainty that may not always allow for complete prediction and control (Prasad, 2017). Given that the role of the researcher involved the behavioural data collection of consumers and the interpretation thereof within a societal setting, a post-positivist paradigm was deemed to be appropriate for the investigation.

In alignment with the research paradigm, the study followed a quantitative approach whereby the researcher made use of a causal research design as the investigation primarily concerned gaining an understanding of the relationship between two interconnected variables. The design of the study, therefore, included data collection via the distribution of questionnaires followed by descriptive and inferential statistical analyses, including an exploratory factor analysis and a regression analysis.

3.2 Ethical considerations

Various issues surrounding consent, harm, permission, personal information, confidentiality and anonymity were identified prior to the research investigation taking place and were addressed by the researcher. Permission to conduct the investigation was thus sought from the University of Cape Town (UCT) Higher Degrees Committee, the UCT Commerce Faculty Ethics in Research Committee and each of the shopping centres from which the researcher conducted data collection activity. In addition, participants were required to provide their written consent for the usage of their responses within the confines of the study.

3.3 Target population and sampling approach

The target population included all emergent black South African middle-class consumers (Zikmund & Babin, 2017). The total number of these consumers accounted for approximately 1.5 million individuals, of which an estimated 765 000 were female, i.e., 51 percent (StatsSA, 2011). Dual sampling techniques (convenience and snowball sampling) were chosen in regard to this investigation for two primary purposes. Firstly, the researcher was required to conduct the research within a specified time frame. Thus, convenience sampling permitted the investigator to locate respondents with ease and in a cost-effective way. Secondly, as the researcher did not have a large network of EBMC women in Durban, making use of snowball sampling ensured that a greater number of respondents were able to be located through the referral process.

Malhotra (2017) alludes to the fact that a sampling frame serves as a representation of an investigation's target population. This frame, therefore, assists in identifying a sample. However, as this investigation used non-probability sampling strategies, a known probability of respondent selection was not possible. The study did, however, yield a sample size of 350 EBMC female respondents located in five key pre-selected Durban suburbs. As the target population (emergent black South African middle-class women in Durban) consisted of a very broad range of individuals, the sample was further specified on an age, location and income basis.

The researcher, therefore, targeted individuals aged between 20 and 29 years who resided in one of five suburbs within the Durban city area (either Kloof, Pinetown, Newlands East, Newlands West or Westville) and who earned a monthly income of between R6 000 to R12 000. These suburbs were specifically targeted as they were regarded to be predominantly black middle-class and were likely to include individuals categorised as 'emerging' rather than 'established' middle class (StatsSA, 2017). Table 1 highlights the locations within Durban where data collection was conducted.

Of the 350 distributed surveys, 301 respondents indicated they were willing to participate and thus completed the questionnaire. In order to establish whether this reduced sample size was deemed to be adequate enough to perform a factor analysis, a Kaiser-Mayer-Olkin (KMO) test was performed using SPSS. Typically, KMO values greater than 0.6 indicate that the size of the sample is sufficient (Gordon, 2015). The analysis revealed a KMO measure of 0.873, thereby confirming the number of surveyed individuals as being satisfactory.

3.4 Research instrument, hypothesis formulation and data analysis

The data collection for this study took place in five selected suburbs located in Durban, namely; Kloof, Pinetown, Newlands East, Newlands West and Westville. More specifically, data was collected within high-traffic areas (including shopping malls and transport hubs) to attract a multitude of respondents. The data collection preparation process began with what are regarded to be the preliminary requisites of methodology, namely, the questionnaire design and the pretest (Malhotra, 2015). The pre-test was conducted with 5 percent of the targeted number of respondents before the questionnaire was finalised and distributed on a larger scale. The measurement instrument comprised 40 questions and was divided into three segments (A: screening questions, B: attitudinal antecedents, i.e., the measured independent variables and C: purchase decision-making). Section A of the questionnaire centred on questions which allowed for the researcher to distinguish whether or not the respondent fell within the targeted sample group and included four 'yes/no/prefer not to answer' format statements. Thus, if the respondent answered 'no' to any of the questions within this segment, they were requested to discontinue completing the questionnaire. However, if the respondent indicated otherwise, they were encouraged to complete the remaining sections of the questionnaire (sections B and C). These sections related to the data collection of the six attitudinal antecedents included in the study's conceptual framework through the usage of 7-point Likert scales. Participants thus indicated their level of agreement with statements by selecting a number from 1 to 7, ranging from 'entirely agree' coded as 1, to 'entirely disagree', coded as 7. Questions pertaining to each of the attitudinal variables were negatively or positively phrased. Table 2 outlines the hypotheses of the investigation, their relevant factor grouping per the conceptual framework and an indication of the way in which questions relating to each antecedent were framed.

Two quantitative research software tools were utilised for the data interpretation process. Once each questionnaire was numbered, the researcher coded the respondents' responses into a Microsoft Excel spreadsheet. The coding of each response coincided with that of the numbers apparent on the Likert scale, apart from those included in the screening segment of the questionnaire, where 1=yes, 2=no and 3=prefer not to answer. Once the researcher had finalised the coding process, the coded values were inputted into the statistical software programme SPSS.

In order to primarily assess the reliability of the data stemming from the research instrument, the researcher calculated Cronbach's Alpha value per variable. Thereafter, descriptive and inferential statistical analyses were performed. In terms of descriptive statistics, the research sought to uncover the mean and standard deviation per item. With regard to inferential statistics, the researcher utilised an exploratory factor analysis and a regression analysis whereby information pertaining to the variables' Eigenvalues, percentage of variance, factor loadings and p-values were uncovered. In order to determine how many factors should have been considered, the number of eigenvalues greater than 1 were primarily examined, which accounted for a fairly large proportion of the variation. Thus, in cases where only one eigenvalue appeared to be greater than 1, one factor was extracted (Gorsuch, 2014).

Thereafter, an exploratory factor analysis (EFA) was performed. This type of analysis was selected given the attempt to discover the nature of the variables infusing a set of responses, as opposed to a confirmatory factor analysis (CFA), which tests whether the data fit a hypothesised measurement model and is largely theory-driven (Kline, 2014). Furthermore, a regression analysis was utilised for the purposes of the study as this form of analysis ultimately estimates and models relationships among two or more variables (Zikmund & Babin, 2017). In addition, regression analyses have proven to be successful when determining connections between attitudinal variables and consumer-behaviour-related outputs (De Matos et al., 2007; Kumar, 2017). Though multiple forms of regressions exist, every analysis fundamentally explores the way in which independent variables influence the dependent variable (Gordon, 2015). The results of descriptive and inferential statistical analyses are presented hereafter.

 

4. RESULTS AND DISCUSSION

The following segment details the results of the study and provides a discussion regarding the implications of its findings.

4.1 Cronbach's alpha

The following table provides the Cronbach's Alpha values for each of the questionnaire variables defined as knowledge, perceived risk, ethical obligation, product price, product quality, service quality and purchase decision-making.

 

Table 3

 

As shown above, the coefficient values concerning items 2.1 to 2.30 are above 0.8, indicating a high degree of internal consistency. The Cronbach's Alpha value relating to the third and final section on purchase decision-making of the questionnaire appears to be lower, though this value is still deemed to fall within broadly acceptable norms (Leppink, 2019).

4.2 Descriptive statistics

The extraction of responses allowed the researcher to perform descriptive analyses, which are further outlined, per variable, in Table 4.

All items included in the knowledge section were positively phrased. Thus, individuals were required to indicate their response from 'entirely agree' to 'entirely disagree'. With mean values ranging from 2.96 to 3.18, it appears that the majority of participants agreed that they were knowledgeable with regard to sale elements and characteristics of counterfeit luxury-branded handbags. Moreover, the standard deviation values relating to this antecedent were minimal, indicating little dispersion around the mean and, in turn, a high level of consensus among respondents.

The first four items in the perceived risk segment were negatively phrased, while the last item (2.10) was relayed positively. The mean results for questions 2.6 to 2.9 ranged from 4.16 to 4.41, demonstrating that most respondents neither agreed nor disagreed with the statements in relation to perceived risk. Thus, it appears that most participants were unclear about the risk associated with purchasing the handbags.

The mean response to question 2.10, 'I believe that there will be no repercussions for purchasing these bags', however, is stated as being 3.27, indicating that the majority of respondents displayed some agreement with the statement posed. As in the case of the knowledge antecedent, the standard deviation for this variable was not substantive. The mean values for the positive statements regarding ethical obligation were documented to be slightly above 3, portraying that generally, participants somewhat agreed that purchasing counterfeit luxury-branded handbags was not shameful and that there was 'nothing wrong with the act'. The mean values for questions 2.13 to 2.15, on the other hand, ranged from 4.26 to 4.39, displaying that the bulk of individuals neither agreed nor disagreed about feelings of guilt post-purchase, choosing not to purchase the bags on a legal basis and considering the act of purchasing as wrongful. The standard deviation values for these items were also regarded to be low.

The product price segment of the questionnaire consisted of three positive statements (2.16, 2.17, 2.19), and two negatively phrased statements (2.18, 2.20). With reference to product price, the mean values in connection with the positive statements ranged from 2.86 to 3.14, portraying that the majority of individuals somewhat considered the counterfeit bags to be a good value for money and would pay for one, should they receive a fair or discounted price. The mean values associated with the negative statements, however, were based around 4, indicating that respondents neither agreed nor disagreed with statements concerning higher prices. Standard deviation values for price-related were also regarded as being low.

The first four questions, which fell under product quality, were positively phrased, with the final item posed negatively. Mean response values for the positive statements in terms of product quality ranged from 3.06 to 3.22, portraying that most respondents exhibited a degree of agreement in terms of quality serving as an important purchase decision-making factor, as well as the fact that the handbags concerned were generally of good quality, and would serve to be a good investment. The mean value for the final item in this segment (question 2.25), on the other hand, was 4.45, displaying that many respondents neither agreed nor disagreed with purchasing a poor to average-quality counterfeit luxury-branded handbag. In addition, the standard deviation values for items in this segment were not deemed to be significant.

The five items included within service quality were all positively phrased. The mean values for the items relating to service quality centred around 3, indicating that most of the respondents displayed some agreement in relation to good customer service positively affecting their purchase decision. As with the other attitudinal antecedents, standard deviation outputs were minimal.

All six items which fell under attitude to purchase were negatively phrased., The standard deviation among the items pertaining to attitude to purchase was found to be low. While the first four items (relating to knowledge, perceived risk, ethical obligation and product price) exhibited mean values around 4, the last two items (concerning product quality and service quality) provided means of 3.34 and 3.36. Thus, it would appear that the impact of product and service quality more negatively influences consumers' decision to purchase. This evidence, however, cannot be deemed conclusive. As a result, inferential statistical analyses were required to provide a greater sense of clarity on the matter.

4.3 Inferential statistics

In order to draw conclusive results from the data collected, the researcher conducted a series of inferential statistical analyses, including factor and regression analyses. The results of these analyses are presented hereafter.

4.3.1 Factor analysis

Aside from adequate commonalities and component matrix values, a factor analysis extraction is based on eigenvalues (Kline, 2014). These values measure how much of the variance in the original observed variables can be explained by the factor (Jackson, 2018). In order to determine how many factors should be considered, the number of eigenvalues greater than 1 are examined, which account for a fairly large proportion of the variation. Thus, in cases where only one eigenvalue appears greater than 1, one factor is extracted (Gorsuch, 2014).

In order to extract factors for purposes of analysis, it was primarily necessary to identify the number of factors per antecedent. Table 5, therefore, details the eigenvalues above 1 for the items making up each component and, in turn, the number of factors selected per variable.

As depicted in Table 5, one factor per variable was extracted, with the exception of purchase decision-making, which produced two eigenvalues greater than 1. Hence, in this case, two factors were extracted, displaying a cumulative variance of 69.56 percent. A varimax rotation was performed on each factor thereafter, with the factor loadings for purchase decision-making presented in Table 6. This rotation served to transform the original factors into new ones, allowing for an easier interpretation of the factors (Coolican, 2017).

From Table 6, it is apparent that Factor 1 loads higher on the variable's knowledge, perceived risk, ethical obligation and product price. Interestingly, Factor 2, however, is largely aligned with product and service quality. The Varimax rotation, therefore, provides more insight into the relationships evident between the original variables and the factor that is extracted. In order to gain a complete understanding of the relationships evident between the attitudinal antecedents and purchase decision-making, however, regression analyses were required to be performed.

4.3.2 Regression analysis

Regression analysis is deemed to be a powerful statistical method, allowing researchers to examine the relationships between two or more elements of interest (Darlington & Hayes, 2016). Though multiple forms of regressions exist, every analysis fundamentally explores the way in which independent variables influence the dependent variable (Gordon, 2015). In this instance, two approaches to regression-based statistical modelling were considered. The first regression model approach was based on the factor scores obtained from the factor analysis performed on the variables listed in the factor analysis. The factor scores for attitude to purchase thus served as the dependent variable, with the factor scores for knowledge, perceived risk, ethical obligation, product price, product quality and service quality making up the independent variables. As mentioned earlier, two factors were apparent in terms of the attitude to purchase variable. Thus, each of these factors are used as the dependent variable in two separate regression models presented hereafter.

The second statistical approach, on the other hand, was founded on obtaining the total scores of each variable. In cases where negative and positive questions were posed, certain items were recoded to reflect the correct summation of the total. The attitude to purchase total was, therefore, deemed to be the dependent variable in the regression model, with the total scores for the attitudinal antecedents serving as the independent variables.

The regression analysis using factor scores was the natural progression following the factor analysis. Two regression models were fitted, each associated with one of the two attitudes to purchase factors. The results of this analysis are presented in Tables 7 and 8.

In Table 7, it appears that the p-values for perceived risk, ethical obligation and product price are minimal and less than 0.05. Draper and Smith (2014) explain that p-values less than 0.05 are indicative of the corresponding variables being significant predictors of the dependent variable in the model. Table 6 depicts that Factor 1 loads highly on the variables' knowledge, perceived risk, ethical obligation and product price'. The current regression model illustrates only 'perceived risk, ethical obligation and product price' as a significant attitude to purchase predictor variables. Knowledge exhibits a p-value of 0.38, that is, one greater than 0.05, and is thus not recognised as a significant predictor variable within this dimension of attitude to purchase. As depicted earlier in Table 7, the attitude to purchase Factor 2 loadings weighed heavily on product quality and service quality. In Table 8, it is evident that knowledge, product price and service quality employ p-values less than 0.05, thus serving to be significant predictors of the Factor 2 regression model.

In order to perform this regression analysis, a series of totals was created by adding the responses to items per questionnaire variable. As the responses to items were based on a Likert scale, which ranged from numbers 1 to 7, the minimum total was stated as 5; that is, if a respondent selected 1 for all five questions. The maximum total, on the other hand, was 35 if a respondent were to have selected 7 for each of the five questions. In terms of the final section of the survey (Part 3), however, which comprised six questions, the minimum total was 6 and the maximum 42. The coefficients produced by regression analysis using this approach are detailed in Table 9 hereafter.

The totals derived from the above regression analysis indicate that four antecedents display p-values of less than 0.05. This, therefore, indicates a significant predictive influence of the attitudinal antecedents, perceived risk, ethical obligation, product price and service quality on attitude to purchase. In addition, it may be deduced that knowledge and product quality, both displaying p-values greater than 0.05, cannot be regarded as substantially influential attitudinal factors within the attitude to purchase context.

4.4 Assessing findings within the context of existing studies

Per Table 10 below, the majority of the hypotheses failed to be rejected. Certain results yielded similarities to past attitude behavioural-based investigations, while others provided alternate evidence. In some cases, novel consumer insight was generated.

Perceived risk and ethical obligation have been documented to consistently influence the consumer decision-making process in a multitude of research investigations (Bhatia, 2017; Marticotte & Arcand, 2017). Thus, the results pertaining to the study at hand reinforce this phenomenon, with both factors producing a high degree of significance. Existing research relating to the influence of other attitude factors, such as knowledge, however, was found to be less consistent, with several authors having documented varying results (Alfadl, 2017; Kozar & Stehl, 2016). As such, while the significance of knowledge supports the arguments of researchers such as Michealidou and Christodoulides (2011) and Marcketti and Shelley (2009), they contradict the investigative outcomes of others (Kozar & Stehl, 2016; Alfadl, 2017). On account of these discrepancies, further research efforts may be required in order to provide a greater degree of consensus regarding these three antecedents in particular.

The variables product price and product quality (derived from the Attitude Towards the Object model) are absent in many counterfeit-related investigations (Prakash & Pathak, 2017). Additional consumer behaviour research, however, contends that customers are often attracted to low-price offerings, with the concept of 'value for money' also serving a major role in the attitude formation and consumption process (Kolar, 2014). Interestingly, while the results of the study maintain the importance of price in terms of building purchase intent, they also indicate that respondents did not view the quality of the goods concerned as an integral factor regarding their consumption thought process.

While the attitudinal factor of service quality was considered to be a novel integration in the framework, the significance of the antecedent served to support the findings and general sentiment surrounding the positive correlation evident between customer service and subsequent purchase behaviour among South African consumers (Nielson, 2015; Shah et al., 2017).

 

5. DISCUSSION

The study primarily sought to uncover the influence of factors on attitude to purchase in relation to luxury counterfeit handbags. Of the results yielded from the investigation, some antecedents appeared to support existing research, while others produced varying results. Producing a high degree of significance, the antecedents perceived risk and ethical obligation reinforce the findings of several investigations (Bhatia, 2017; Marticotte & Arcand. 2017). Existing research relating to the influence of other attitude factors, such as knowledge, however, is deemed to be less consistent, with authors having documented varying results (Kozar & Stehl, 2016; Alfadl, 2017). Thus, while the significance of knowledge supports the arguments of researchers such as Michealidou and Christodoulides (2011) and Marcketti and Shelley (2009), they contradict the investigative outcomes of others (Kozar & Stehl, 2016; Alfadl, 2017). Given these discrepancies, further research is required in order to establish the degree of similarity between the results of this study and other findings within both a South African and global context.

Although the variables of the product price and product quality are absent in many counterfeit-related investigations, consumer behaviour research contends that customers are often attracted to low-price offerings, with the concept of 'value for money' also serving a major role in the attitude formation and consumption process (Kolar, 2014; Prakash & Pathak, 2017). Interestingly, while the results of the study maintain the importance of price in terms of building purchase intent, they also indicated that respondents did not view the quality of the goods concerned as an integral factor regarding their consumption thought process.

While the attitudinal factor of service quality was deemed as a novel integration in the framework, the significance of the antecedent served to support the findings and general sentiment surrounding the positive correlation evident between customer service and subsequent purchase behaviour among South African consumers (Nielson, 2015; Shah et al., 2017).

Overall, the investigation ultimately confirmed Hypothesis 1, i.e., that IPS (Individual, Product and Service) factors do have an influence on the attitude to purchase of counterfeit luxury handbags.

The study thus provides an important contribution in terms of illustrating that the demand dimension of curbing counterfeit purchases is as integral, if not potentially more significant, than addressing the challenge from the supply perspective. Furthermore, the investigation displays that of the 6 factors tested, 5 impacted heavily on attitude to purchase within a local context. While the antecedents included in the study's framework may not demonstrate the impact of all contributing components regarding attitude to purchase, they provide the foundation from which South African law enforcement agencies and anti-counterfeiting bodies can begin to focus their efforts.

While some factors, such as product price, product quality and quality service, are more aligned to that of sellers and manufacturers of counterfeit goods, perceived risk, ethical obligation and knowledge are regarded to be factors that hold the potential to be elevated within a demand context. Mass educational campaigns surrounding counterfeit purchasing, for example, could drive awareness of the illegal nature of counterfeit goods whilst simultaneously raising the ethical concerns of counterfeit purchases. In addition, the perceived risk of buying counterfeit items may be elevated by increasing the number of law enforcement officials in regions where counterfeit sales take place. Moreover, communication issued by South African policing bodies could convey that the act of counterfeit purchase supports an illegal practice, amplifying the associated risk. Thus, attempting to curb counterfeit purchases with a consumer-centric strategy may serve to not only decrease the demand for counterfeit products but ultimately reduce the global market for these goods along with the lure of profit, which primarily enables the cycle of counterfeit production.

 

6. MANAGERIAL IMPLICATIONS

The results of this research investigation fundamentally confirm the importance of marketers and law enforcement officials alike understanding and tackling the issue of counterfeit luxury handbag sales from the demand or consumer side of the problem. The data collected provides insight into the fact that attitude influences consumers' decision to purchase, but also details which factors buyers of these goods consider to be more significant than others. As such, it allows those combating counterfeit sale activity to focus their efforts and resources in a predetermined manner. To date, for example, most anti-counterfeit campaigns have focused on the education of consumers in an attempt to create awareness surrounding the illegality of the sale and purchase of these goods.

Evidence stemming from this investigation strengthens the argument towards this approach, portraying that knowledgeable consumers are more deterred from making counterfeit purchases than those who do not display such a degree of awareness. Therefore, it may be valuable to consider additional efforts aligned with combating counterfeit sales through an increased consumer consciousness concerning the true nature of the products.

Though officials may have little control over elements such as product quality, product price and service quality, they are able to exert some influence in areas concerning perceived risk and ethical obligation. Law enforcement officials, for example, could make more of a concerted effort to display their presence in or around sales areas. Moreover, marketers are able to specifically direct campaigns towards the possible repercussions associated with counterfeit handbag purchases, as well as placing emphasis on the shameful, guilt-associated repercussions of illegal consumption.

 

7. CONCLUSION

Understanding the factors that drive consumers to make purchase decisions has historically been identified as one of the major focus areas within consumer behaviour research. While gaining this knowledge under ordinary purchasing contexts remains important, obtaining an idea of the drivers of purchase within illegal parameters holds the added benefit of potentially deterring criminal activity. Thus, the results of this study provide for interesting learnings.

The factor and regression analyses ultimately depicted that five of the six attitudinal antecedents significantly impacted attitude to purchase, namely, knowledge, ethical obligation, perceived risk, product price and service quality. Product quality, therefore, served to be the sole variable not substantially influencing the consumers' attitude to purchase. Nonetheless, it must be noted that product quality does influence attitude to purchase to a limited degree. Thus, of the hypotheses, H₁, H₂, H₃, H₄, H₅ and H₇ failed to be rejected.

Although the researcher encountered some limitations throughout the research process, evidence from this study principally confirms that IPS factors do, in fact, influence the attitude to the purchase of Durban's emergent black middle-class women in relation to luxury-branded handbags. While the product quality of these handbags failed to generate a high degree of impact on the consumers' attitude to purchase, this finding, too, provides insight into consumer behaviour.

Therefore, this investigation provides a valuable contribution to expanding the knowledge base regarding attitudes to purchasing within a counterfeit environment. Moreover, it contributes to the body of literature relating to attitude to purchase factors, with the majority of factors examined signalling significant influence. Furthermore, the results of the study provide the basis for directing anti-counterfeit purchase communication and the broader anti-counterfeit consumer-focused response.

Future research may investigate the degree to which additional factors, such as values, significant others, and culture, impacts attitude to purchase within this context. In addition, studies focusing on other counterfeit goods, for example, clothing or jewellery, could provide an interesting comparison in terms of influential attitude to purchase factors. Aside from contrasting the response to the items under investigation, future research efforts could pertain to different consumer groups, such as the established black middle class. Limitations of the study mainly existed in terms of geographic reach, given that the researcher was based in Durban. It is therefore recommended that forthcoming research investigations collect data from a wider audience in order to draw more generalisable findings relating to the South African population. A more detailed account of the study's limitations and areas for future studies are detailed hereafter.

 

8. LIMITATIONS AND FUTURE STUDIES

While the investigator was able to successfully execute on the research process and provide for conclusive results, certain limitations preceded the study. Firstly, from a methodology standpoint, the researcher chose to make use of non-probability sampling, which according to Malhotra (2017), relates to the non-random selection of a sample. In other words, all members of a population do not possess the same probability of selection. Thus, evidence stemming from these samples cannot be generalised in terms of the entire population. The researcher's actualised sample size of 301 respondents, however, indicates that meaningful conclusions may be drawn. Although, if possible, it is recommended that researchers attempt to increase the number of participants in future studies.

Aside from the sampling approach, the researcher was limited in terms of geographic flexibility and in turn, reach. Thus, the study solely pertained to participants located in Durban, KwaZulu-Natal, the data of which, therefore, cannot be extended to national level. In addition, a specific age range of women was targeted, thus excluding data from participants whose data may have altered the results of the investigation. Hence, it is recommended that researchers collect data from multiple provinces as well as increase the age range of participants in future research investigations. Lastly, the researcher was often limited due to situational factors arising from shopping mall environments. With many consumers visiting malls during their lunch breaks or after work, time proved to be an inhibiting factor. Thus, it is recommended that researchers investigate alternate methods of data collection in the future.

On the other hand, the results of the study also provide the foundation for opportunities for interesting future research. Firstly, the study confirms the significance of service quality in relation to attitude formation and, subsequently, attitude to purchase. Thus, it may be interesting for researchers to determine whether this antecedent continues to play an important role across varying South African provinces, as well as in international settings. Secondly, upcoming research could potentially focus on uncovering the reasoning behind the lack of significance of product quality of members of the EBMC in relation to counterfeit handbags. Thirdly, subsequent investigative work could focus on determining whether a difference is evident concerning the attitudes of the established segment of the South African black middle class in comparison with the results evident from this study. Lastly, further research efforts may be initiated following to the commencement of campaigns created alongside the results of the study. Thus, investigators will be able to more accurately assess whether the practical implications of the study, do, in fact, alter the behaviour of consumers.

 

REFERENCES

Adelman, M., Berry, M. & Coles, H. 2014. Shipping first drafts: a collection of travel essays. Morrisville: Lulu Press        [ Links ]

Abarracin, D. & Johnson, B.T. 2018. The handbook of attitudes, volume 1: basic principles. 2^nd ed. New York: Routledge. [https://doi.org/10.4324/9781315178103].         [ Links ]

Alexander, P., Ceruti, C. & Motseke, K. 2013. Class in Soweto. Durban: University of KwaZulu-Natal Press.         [ Links ]

Alfadl, A.A. 2017. Perspective, knowledge, attitude and belief of various stakeholders on medicines quality: counterfeit and substandard medicines. London: Academic Press. 345-360. [https://doi.org/10.1016/B978-0-12-8112281.00021-2].         [ Links ]

Amaral, N.B. 2016. Viewing usage of counterfeit luxury goods: social identity and social hierarchy effects on dilution and enhancement of genuine luxury brands. Journal of Consumer Psychology, 26(4):483-495. [https://doi.org/10.1016/jJcps.2016.02.004].         [ Links ]

Atwal, G. & Bryson, D. 2017. Luxury brands in China and India. Singapore: Springer. [https://doi.org/10.1057/978-1-137-54715-6].         [ Links ]

Babin, B.J & Harris, E. 2015. Consumer Behaviour. Boston: Cengage Learning        [ Links ]

Bagozzi, R., Gurhan-Canli, Z. & Priester, J. 2002. The social psychology of consumer behaviour. London: McGraw-Hill Education.         [ Links ]

Barkway, P. 2013. Psychology for health professionals. 2^nd ed. Chatswood: Elsevier Publishing.         [ Links ]

Beckert, J. & Musselin, C. 2013. Constructing quality: the classification of goods in markets. Oxford: Oxford University Press. [https://doi.org/10.1093/acprof:oso/9780199677573.001.0001].         [ Links ]

Bell, B. 2016. BBC News: what's wrong with buying fake luxury goods? [Internet: https://www.bbc.com/news/uk-england-36782724?tblang=english; downloaded on 26 November 2017]        [ Links ]

Bhatia, V. 2017. Examining consumers' attitude towards purchase of counterfeit fashion products. Journal of Indian Business Research, 10(2):193-207. [https://doi.org/10.1108/JIBR-10-2017-0177].         [ Links ]

Brennan, L., Binney, W. & Parker, L. 2014. Social marketing and behaviour change: models, theory and applications. Cheltenham: Edward Elgar. [https://doi.org/10.4337/9781782548157].         [ Links ]

Burger, R., Louw, M., Pegado, B.B.I.D. & Van Der Berg, S. 2014. Understanding consumption patterns of the established and emerging South African black middle class. Stellenbosch Economic Working Paper. Stellenbosch: Stellenbosch University. [https://doi.org/10.2139/ssrn.2693652].         [ Links ]

Choi, T. & Shen, B. 2016. Luxury fashion retail management. Singapore: Springer. [https://doi.org/10.1007/978-981-10-2976-9].         [ Links ]

Chiu, W. & Leng, H.K. 2016. Consumers' intention to purchase counterfeit sporting goods in Singapore and Taiwan. Asia Pacific Journal of Marketing and Logistics, 28(1):23-36. [https://doi.org/10.1108/APJML-02-2015-0031].         [ Links ]

Coolican, H. 2017. Research methods and statistics in psychology. 6^th ed. New York: Psychology Press. [https://doi.org/10.4324/9780203769836].         [ Links ]

Darlington, R.B. & Hayes, A.F. 2016. Regression analysis and linear models. London: Guilford Press.         [ Links ]

De Matos, C.A., Ituassu, C.T. & Rossi, C.A. 2007. Consumer attitudes towards counterfeits: a review and extension. Journal of Consumer Marketing, 24(1):36-47. [https://doi.org/10.1108/07363760710720975].         [ Links ]

Draper, N.R. & Smith, H. 2014. Applied regression analysis. 3^rd ed. New York: Wiley and Sons.         [ Links ]

Fabrizio, M. 2016. Global marketing strategies for the promotion of luxury goods. Philadelphia: IGI Global.         [ Links ]

Frederickson, G.H. & Ghere, R.K. 2013. Ethics in public management. 2^nd ed. New York: Routledge. [https://doi.org/10.4324/9781315704517].         [ Links ]

Freer, A. 2018. The accessory handbook. 1^st ed. Berkeley: Ten Speed Press.         [ Links ]

Gordan, R.A. 2015. Regression analysis for the social sciences. 2^nd ed. New York: Routledge.         [ Links ]

Gorsuch, R.L. 2014. Factor analysis. Classic ed. New York: Routledge. [https://doi.org/10.4324/9781315735740].         [ Links ]

Hanssens, D.M., Pauwels, K.H., Srinivasan, S., Vanhuele, M., & Yildirim, G. 2014. Consumer attitude metrics for guiding marketing mix decisions. Marketing Science, 33(4):534-550. [https://doi.org/10.1287/mksc.2013.0841].         [ Links ]

Howell, K.E. 2012. An introduction to the philosophy of methodology. 1^st ed. London: Sage. [https://doi.org/10.4135/9781473957633].         [ Links ]

Hung, Y., De Kok, T.M. & Verbeke, W. 2016. Consumer attitude and purchase intention towards processed meat products with natural compounds and a reduced level of nitrite. Meat Science, 121:119-126. [https://doi.org/10.1016/j.meatsci.2016.06.002].         [ Links ]

International Anti-Counterfeiting Coalition (IACC). 2017. Counterfeiting adds up. [Internet: https://www.iacc.org/resources/about/statistics; downloaded on 26 November 2017].         [ Links ]

Iqani, M. 2015. Agency and affordability: being black and 'middle class' in South Africa. South-North Cultural and Media Studies, 29(2):126-145. [https://doi.org/10.1080/02560046.2015.1039200].         [ Links ]

Jackson, E.A. 2018. Research methodology in the social sciences: perspectives on Sierra Leone. Morrisville: Lulu Publishing.         [ Links ]

Kapferer, J. 2015. Kapferer on luxury: How luxury brands can grow yet remain rare. New York: Kogan Page.         [ Links ]

Khunou, G. 2017. South Africa's emergent middle class. New York: Routledge. [https://doi.org/10.4324/9781315660332].         [ Links ]

Kline, P. 2014. An easy guide to factor analysis. 1^st ed. New York: Routledge. [https://doi.org/10.4324/9781315788135].         [ Links ]

Kolar, T. 2014. The effect of general consumer attitudes and personality traits on attitudes towards domestic and multinational products in the Balkan region. Journal of East European Management Studies, 19(1):58-80. [https://doi.org/10.5771/0949-6181-2014-1-58].         [ Links ]

Kotze, H., Du Toit, P., Khurou, G. & Steenkamp, C.L. 2016. Emergent middle class in South Africa. [Internet: http://resep.sun.ac.za/index.php/projects/emergent-middle-class-in-south-africa/; downloaded on 15 October 2017].         [ Links ]

Kozar, J.M. & Stehl, K. 2016. Consumer knowledge, attitudes, and purchase intentions of counterfeit fashion goods: an initiative to curbing consumer demand in the marketplace. Cham: Springer. 143-154. [https://doi.org/10.1007/978-3-319-26647-327].         [ Links ]

Kumar, R. 2017. Research methodology: a step by step guide for beginners. 4^th ed. Thousand Oaks: Sage.         [ Links ]

Lee, M. 2014. Africa's world trade: informal economies and globalisation from below. London: Zed Books. [https://doi.org/10.5040/9781350218253].         [ Links ]

Leppink, J. 2019. Statistical methods for experimental research in education and psychology. Switzerland: Springer International Publishing. [https://doi.org/10.1007/978-3-030-21241-4].         [ Links ]

Letsch, C. 2011. Turkey cracks down on counterfeit goods. [Internet: https://www.theguardian.com/world/2011/oct/17/turkey-counterfeit-goods-trade; downloaded on 18 June 2018].         [ Links ]

Malhotra, N.K. 2015. Essentials of marketing research: a hands-on orientation. 1^st ed. London: Pearson Education.         [ Links ]

Malhotra, N.K. 2017. Marketing research: An applied orientation. Upper Saddle River: Pearson Education.         [ Links ]

Malhotra, N.K., Nunan, D. & Birks, D.F. 2017. Marketing research: an applied approach. 5^th ed. London: Pearson Education. [https://doi.org/10.4324/9781315088754].         [ Links ]

Marcketti, S.B. & Shelley, M.C. 2009. Consumer concern, knowledge and attitude towards counterfeit apparel products. International Journal of Consumer Studies, 33(3):327-337. [https://doi.org/10.1111/j.1470-6431.2009.00748.x].         [ Links ]

Marticotte, F. & Arcand, M. 2017. Schadenfreude, attitude and the purchase intentions of a counterfeit luxury brand. Journal of Business Research, 77:175-183. [https://doi.org/10.1016/j.jbusres.2016.12.010].         [ Links ]

Mattes, R. 2014. South Africa's emerging black middle class: a harbinger of political change? Journal of International Development, 27(5):665-692. [https://doi.org/10.1002/jid.3100].         [ Links ]

McNeil, P. & Riello, G. 2016. Luxury: a rich history. Oxford: Oxford University Press.         [ Links ]

Meyer, N. 2019. South African female entrepreneurs' business styles and its influence on various entrepreneurial factors. Forum Scientiae Oeconomia, 7(2):25-35. [https://doi.org/10.23762/fSo_vol7_no2_2].         [ Links ]

Michealidou, N. & Christodoulides, G. 2011. Antecedents of attitude and intention towards counterfeit symbolic and experiential products. Journal of Marketing Management, 27(9):976-991. [https://doi.org/10.1080/0267257X.2010.549189].         [ Links ]

Nair, Y. 2012. R60 million swoop at Durban harbour. [Internet: https://www.iol.co.za/news/south-africa/r60m-swoop-at-durban-harbour-1206863; downloaded on 26 December 2017].         [ Links ]

Ncube M. & Lufumpa, C.L. 2015. The emerging middle class in Africa. New York: Routledge. [https://doi.org/10.4324/9781315757896].         [ Links ]

Nielsen. 2015. Global corporate sustainability report. [Internet: http://www.nielsen.com/content/dam/nielsenglobal/eu/docs/pdf/Global%20Sustainability%20Report%20DIGITAL%20FINAL.pdf; downloaded on 27 December 2017].         [ Links ]

Nxumalo, M. & Mthethewa, N. 2016. Hawks bust counterfeit operation in KZN. [Internet: https://www.iol.co.za/news/south-africa/kwazulu-natal/hawks-bust-nike-counterfeit-operation-in-kzn-2082486; downloaded on 26 December 2017].         [ Links ]

Organization of Economic Co-operation and Development (OECD). 2016. Trade in Counterfeit and Pirated Goods: Mapping the economic impact [Internet: http://www.oecd.org/governance/trade-in-counterfeit-and-pirated-goods-9789264252653-en.htm; downloaded on 1 June 2018].         [ Links ]

Paquette, D. 2018. Asia and Pacific: Trump's trade war brings unexpected boom for knockoff designer bags from China [Internet: https://www.washingtonpost.com/world/asiapacific/an-unexpected-winner-from-trumps-trade-war-knockoff-designer-bags-from-china/2018/09/19/69d8fd80-b667-11e8-b79f-f6e31e555258story.html; downloaded on 1 October 2018].         [ Links ]

Pecotich, A. & Shultz, C.J. 2016. Handbook of markets and economies. New York: Routledge.         [ Links ]

Pohlman, J. & Day, A. 2013. CNBC: behind the billion-dollar counterfeit bag market. [Internet: https://www.cnbc.com/id/100931854; downloaded on 17 June 2018].         [ Links ]

Prakash, G. & Pathak, P. 2017. Determinants of counterfeit purchase: a study on young consumers in India. Journal of Scientific and Industrial Research, 76:208-211.         [ Links ]

Prasad, P. 2017. Crafting qualitative research: beyond positivist traditions. 2^nd ed. New York: Taylor and Francis. [https://doi.org/10.4324/9781315715070].         [ Links ]

Rathore, V. 2013. Luxury brands like Hermes, Gucci & others to take on faster growing fakes in India. The Economic Times. [Internet: http://articles.economictimes.indiatimes.com/2013-08-28/news/41538692_1_luxury-products-luxury-brands-hermes/2; downloaded on 27 December 2017].         [ Links ]

Raustiala, K. & Springman, C. 2012. The knockoff economy: how imitation sparks innovation. Oxford: Oxford University Press. [https://doi.org/10.1093/acprof:osobl/9780195399783.001.0001].         [ Links ]

Schiffman, L.G. & Kanuk, L.L. 2004. Consumer behaviour. 8^th ed. Upper Saddle River: Prentice Hall.         [ Links ]

Schwartz, M. S. 2016. Ethical decision-making theory: an integrated approach. Journal of Business Ethics, 139(4):755-776. [https://doi.org/10.1007/s10551-015-2886-8].         [ Links ]

Shah, M., Tanveer, N. & Azam, Z. 2017. Service quality, customer attitude and subscription behaviour: a structural equation modelling approach to determine mediating relationship. African Journal of Basic and Applied Sciences, 9(1):7-18.         [ Links ]

Shams, K. 2015. As Louis Vuitton knows all too well, counterfeiting is a costly bargain. [Internet: https://www.forbes.com/sites/realspin/2015/06/25/as-louis-vuitton-knows-all-too-well-counterfeiting-is-a-costly-bargain/?sh=5d96373260d3; downloaded on 27 December 2017].         [ Links ]

Shapiro, J.P. & Stefkovich, J.A. 2016. Ethical leadership and decision making in education: applying theoretical perspectives to complex dilemmas. 4^th ed. Routledge: New York.         [ Links ]

Simpson, J. & Lappeman, J. 2017. Marketing in South Africa consumer landscape: concepts and cases. 4^th ed. Pretoria: Van Schaik.         [ Links ]

Som, A. & Blanckaert, C. 2015. Luxury: Concepts, facts, markets and strategies. Singapore: John Wiley and Sons.         [ Links ]

Southhall, R. 2016. The new black middle class in South Africa. Johannesburg: Jacana Media.         [ Links ]

Statistics South Africa (STATSSA). 2011. Census 2011. [Internet: http://www.statssa.gov.za/census/census2011/census products/Census_2011_Key_results.pdf; downloaded on 4 September 2017].         [ Links ]

Statistics South Africa (STATSSA). 2017. Mid-year population estimates [Internet: http://www.statssa.gov.za/publications/P0302/P03022017.pdf; downloaded on 4 September 2017].         [ Links ]

Stewart, B. 2014. How to spot a fake designer handbag: stop being ripped off by scammers. 1^st ed. Durham: Bull City Publishing.         [ Links ]

Stottinger, B. & Penz, E. 2017. Why buy the original, if the fake can do the same? (Can it?): the demand for counterfeit sports gear in an emerging market. the customer is not always right? marketing orientations in a dynamic business world. developments in marketing science: Proceedings of the Academy of Marketing Science. New York: Springer. [https://doi.org/10.1007/978-3-319-50008-9 10].         [ Links ]

Sun, H., Beebe, B. & Sunder, M. 2015. The luxury economy and intellectual property: critical reflections. Oxford: Oxford University Press. [https://doi.org/10.1093/acprof:oso/9780199335701.001.0001].         [ Links ]

Thomas, D. 2015. Deluxe: how luxury lost its lustre. 1^st ed. London: Penguin Books.         [ Links ]

Tomshinsky, I. 2016. Bags & purses: the story of chic and practicality. 1^st ed. Bloomington: Xlibris.         [ Links ]

Waring, A. 2013. Corporate risk and governance: an end to mismanagement, tunnel vison and quackery. 1^st ed. New York: Routledge.         [ Links ]

Willet, M. 2015. Here's the hierarchy of luxury brands around the world [Internet: https://www.independent.co.uk/news/business/here-s-the-hierarchy-of-luxury-brands-around-the-world-a8019126.html; downloaded on 27 December 2017].         [ Links ]

Wittmer, D.P. 2016. Developing a behavioural model for ethical decision making in organizations: conceptual and empirical research. 1^st ed. New York: Routledge.         [ Links ]

Yang, F. 2015. Faked in China: nation branding, counterfeit culture and globalization. 1^st ed. Bloomington: Indiana University Press.         [ Links ]

Zikmund, W.G & Babin, B.J. 2017. Exploring marketing research. 10^th ed. Beijing: Cengage Learning.         [ Links ]