1 Consumer and Product-Related Factors Impacting on Green Product Purchase Behaviour Nombulelo Dilotsotlhe 1599140 Thesis Submitted in fulfilment of the requirements for the degree of Doctor in Philosophy in Marketing Management School of Economic and Business Sciences at the University of the Witwatersrand Supervisor: Dr Helen Duh 2018 i DECLARATION I, Nombulelo Dilotsotlhe, do hereby declare that this thesis is my own, unaided work. It is being submitted for the Degree of Doctor of Philosophy in the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination in any other University. November 2018 ii ACKNOWLEDGEMENTS I would like to express my heartfelt gratitude to the following people for their contribution and assistance towards the completion of this study:  Firstly, my Heavenly Father, for giving me the will-power and perseverance in achieving my goals.  My husband Mokaedi and our two sons Thatoyaone and Masedi, for your love, understanding, support and encouragement. You believed in me when at times, I did not even believe in myself – forever thankful for having you in my life!  My mother Thembalihle Nkwanyana, for your unending support and for listening to my complaints and frustrations.  My research supervisor Dr Helen Duh, for guiding and mentoring me during this process. I am deeply grateful for your encouragement, support and the enthusiasm that you showed for my study topic from the very beginning of this journey. Words cannot express how thankful I am for having had you as my supervisor.  To the University of Johannesburg and WRSETA for financial support. Your contributions made my life much easier, as I did not (at any day) have to worry about funding the costs of this study.  Prof De Meyer-Heydenrych in the Department of Marketing at the University of Johannesburg, for approving my request to study at the University of the Witwatersrand.  Elrina Calitz and team from GfK (South Africa) for assisting me with data collection. Your hard work and patience are truly appreciated.  Dr Maduku, for his specialised support and assistance with the data analysis and interpretation. Your patience in explaining the key concepts of the structural equation modelling and your availability when needed is greatly appreciated.  Lastly, to all my colleagues, friends and Dr Wait, Head of Department (Marketing) at the University of Johannesburg, for your support and encouragement – thank you. iii ABSTRACT Following the growing environmental, health and social concerns about the impact of marketing activities and consumption, consumers are becoming increasingly conscious of the products they use in their households and the food they eat. Even though this awareness is advancing at a slower pace in South Africa when compared to other countries, South African consumers are nonetheless aware of green products and are buying such products. ‘Green products’ can be defined as those products which do not pollute the earth or deplete natural resources and which can be recycled or conserved. Despite the awareness and availability of green products, sales of these products are reportedly low, with an estimated market share of less than 4% worldwide as at 2012. Considering the low demand for green products, there is a clear need to study factors that hinder or help the purchase and consumption of such products. In examining the drivers of green purchase behaviour, the Theory of Planned Behaviour (TPB) has been applied in a variety of pro-environmental, behavioural studies, including research on key factors that influence the intention to consume green products as well as studies on recycling and reducing energy consumption. The findings from these studies have supported the usefulness of the TPB constructs in predicting the intention to go green or to adopt green behaviour. However, the strength of the explanatory power of the TPB model has been widely debated. It has therefore been suggested that more constructs need to be examined in the TPB to improve the variance explained in intention or behaviour. Such constructs could include the different values that consumers enjoy from green products. The Theory of Consumption Values (TCV) has been used to understand such values. However, studies on these values have not examined the product-related factors which help or hinder green product purchase. The Diffusion of Innovation Theory (DOI) addresses product-related factors but has been used chiefly in analysing the adoption of new technologies. Recommendations have therefore been made to test the Diffusion of Innovation Theory in green product purchase behaviour. Upon defining the research problem, this study integrated the TPB, TCV and DOI to understand the consumer-related and product-related factors driving green appliance (e.g. fridges, heaters, freezers) purchase behaviour among black middle class South Africans. This is particularly iv important since studies on green products in South Africa have focused primarily on understanding awareness, perceptions, interest and environmental concerns regarding eco- friendly products or green products and not on how these factors drive green product purchase behaviour. The black middle class was selected because they are considered to be significant in terms of size and spending power and may have the income to purchase green products, which are often considered to be somewhat more expensive. A structured questionnaire was used to collect data from 500 black middle class respondents residing in Gauteng. Confirmatory Factor Analysis and Structural Equation Modelling were the main data analysis methods. Specifically, Partial Least Squares (PLS) Structural Equation Modelling was used to test the hypotheses after testing and obtaining good construct reliability and validity. The results indicated that the integrated conceptual model provided a reasonably good explanation of the consumers’ attitudes towards green appliances as well as their actual purchase and consumption of the products. All the significant independent variables (functional value, conditional value, epistemic value, relative advantage, compatibility and observability) explained 84.5% of the variance in attitude towards green consumption. Attitudes, social value and perceived behavioural control explained 83.1% of the variance in behavioural intention towards green consumption and behavioural intention explained 24.2% of the variance in adoption behaviour. This study made a major contribution to the body of literature on green consumption behaviour by presenting and testing an integrated model with the aim of increasing knowledge on green consumer behaviour. This study also identified key predictors of consumers’ green product consumption behaviour, enabling practitioners to understand which factors influence black middle class consumers in their decision-making regarding green purchase and consumption. This knowledge will help marketing managers design effective strategies to encourage green consumption behaviour among such consumers. Further studies should, however, investigate the factors which would convert the purchase intention into actual behaviour. The study could be replicated among Generation Y consumers, who are also a large and lucrative market segment in South Africa. v TABLE OF CONTENTS Declaration ......................................................................................................................... i Acknowledgements ........................................................................................................... ii Abstract ............................................................................................................................ iii Table of Contents .............................................................................................................. v List of Figures .................................................................................................................. xi List of Tables .................................................................................................................. xii CHAPTER 1: INTRODUCTION AND BACKGROUND OF THE STUDY ........... 1 1.1 INTRODUCTION ...................................................................................................... 1 1.2 PROBLEM STATEMENT ........................................................................................ 3 1.3 OBJECTIVES OF THE STUDY ................................................................................ 5 1.3.1 Primary Objective ............................................................................................. 5 1.3.2 Secondary Objectives ....................................................................................... 6 1.3.2.1 Theoretical Objectives .......................................................................... 6 1.3.2.2 Empirical Objectives ............................................................................ 6 1.4 RESEARCH QUESTIONS ........................................................................................ 7 1.5 PRELIMINARY LITERATURE REVIEW ............................................................... 7 1.5.1 Overview of Green Consumer Behaviour ........................................................ 7 1.5.2 Diffusion of Innovation Theory (DOI) ............................................................. 8 1.5.3 Theory of Planned Behaviour (TPB) .............................................................. 10 1.5.4 Theory of Consumption Values (TCV) .......................................................... 11 1.6 CONCEPTUAL MODEL AND HYPOTHESIS DEVELOPMENT ...................... 12 1.7 OVERVIEW OF RESEARCH METHODOLOGY ................................................. 14 1.7.1 Research Philosophy ....................................................................................... 14 1.7.2 Research Approach ......................................................................................... 14 1.7.3 Research Strategies ......................................................................................... 14 1.7.4 Data Collection Methods or Designs .............................................................. 15 1.7.5 Sampling Design ............................................................................................. 15 1.7.6 Determining the Sample Size ......................................................................... 16 1.7.7 Sampling Procedure ........................................................................................ 16 1.7.8 Questionnaire Design ...................................................................................... 17 vi 1.7.9 Data Analysis .................................................................................................. 17 1.8 CONTRIBUTION OF THE STUDY ....................................................................... 18 1.9 ETHICAL CONSIDERATIONS .............................................................................. 19 1.10 OUTLINE OF THE STUDY .................................................................................. 19 CHAPTER 2: GREEN PRODUCT SUPPLY AND DEMAND GLOBALLY AND IN SOUTH AFRICA: EXAMINING THE BLACK MIDDLE CLASS AS A MARKET SEGMENT.................................................................................................. 21 2.1 INTRODUCTION .................................................................................................... 21 2.2 THE CONCEPT OF GREEN PRODUCTS AND GREEN MARKETING ............ 22 2.2.1 What is a Green Product? ............................................................................... 22 2.2.2 Types and Characteristics of Green Products ................................................. 23 2.2.3 Benefits of Green Product Consumption ........................................................ 27 2.2.4 Defining Green Marketing .............................................................................. 29 2.2.5 Ways in Which a Firm Can Go Green ............................................................ 30 2.3 SUPPLY MARKET OF GREEN PRODUCTS IN SOUTH AFRICA .................... 34 2.3.1 Green Industry in South Africa ....................................................................... 34 2.3.2 Dealers of Green Products in South Africa .................................................... 35 2.3.3 Barriers to the Supply and Marketing of Green Products in South Africa ..... 35 2.4 GREEN PRODUCT DEMAND ............................................................................... 40 2.4.1 Green Product Demand Globally .................................................................... 40 2.4.2 Size of Green Product Demand in South Africa ............................................. 41 2.4.3 Profile of Green Product Consumers ............................................................. 42 2.5 SOUTH AFRICAN BLACK MIDDLE CLASS AS A POTENTIAL MARKET FOR GREEN PRODUCTS .............................................................................................. 43 2.5.1 The Size and Income of the South African Black Middle Class .................... 43 2.5.2 Segments and Consumption Lifestyle of South African Black Middle Class 45 2.5.3 Can the South African Middle Class be an Attractive Market for Green Products? .................................................................................................................. 47 2.6 CONCLUSION ......................................................................................................... 49 vii CHAPTER 3: REVIEW OF DRIVERS OF GREEN PURCHASE BEHAVIOUR 51 3.1 INTRODUCTION .................................................................................................... 51 3.2 GREEN PURCHASE BEHAVIOUR AND ITS REPORTED DRIVERS .............. 51 3.2.1 Defining Green Purchase Behaviour and its Facets........................................ 51 3.2.2 Reported Drivers of Green Purchase Behaviour............................................. 53 3.2.2.1 Demographic Variables ..................................................................... 53 3.2.2.2 Psychographic Factors....................................................................... 56 3.2.2.3 Group Factors .................................................................................... 62 3.2.2.4 Marketing Factors .............................................................................. 66 3.3 CONCLUSION ......................................................................................................... 68 CHAPTER 4: THEORIES AND MODELS EXPLAINING GREEN PRODUCT PURCHASE BEHAVIOUR AND CONCEPTUAL MODEL DEVELOPMENT .. 70 4.1 INTRODUCTION .................................................................................................... 70 4.2 THEORIES AND MODELS OF PURCHASE BEHAVIOUR DRIVERS ............. 70 4.2.1 Theory of Reasoned Action ............................................................................ 71 4.2.2 Theory of Planned Behaviour ......................................................................... 73 4.2.3 Theory of Consumption Values ...................................................................... 75 4.2.4 Diffusion of Innovation Model ....................................................................... 79 4.3 CONSUMER FACTORS DRIVING GREEN PRODUCT ATTITUDE ................. 84 4.3.1 Description of Consumer Factors ................................................................... 84 4.3.2 Functional Value and Green Product Attitude ............................................... 84 4.3.3 Social Value and Green Product Attitude ...................................................... 85 4.3.4 Emotional Value and Green Product Attitude ................................................ 86 4.3.5 Conditional Value and Green Product Attitude .............................................. 87 4.3.6 Epistemic Value and Green Product Attitude ................................................. 87 4.4 PRODUCT FACTORS DRIVING GREEN PRODUCT ATTITUDE .................... 88 4.4.1 Description of Product Factors ....................................................................... 88 4.4.2 Relative Advantage and Green Product Attitude ........................................... 88 4.4.3 Compatibility and Green Product Attitude ..................................................... 89 4.4.4 Complexity and Green Product Attitude ........................................................ 89 4.4.5 Observability and Green Product Attitude...................................................... 90 viii 4.5 GREEN PRODUCT ATTITUDE ............................................................................. 90 4.6 PURCHASE INTENTION AND ACTUAL PURCHASE BEHAVIOUR .............. 91 4.7 CONCEPTUAL MODEL AND HYPOTHESES .................................................... 92 4.7.1 Hypothesised Relationships between Constructs in the Conceptual Model .. 95 4.8 CONCLUSION ......................................................................................................... 99 CHAPTER 5: RESEARCH METHODOLOGY ..................................................... 100 5.1 INTRODUCTION .................................................................................................. 100 5.2 RESEARCH PROCESS ......................................................................................... 100 5.3 RESEARCH PHILOSOPHY .................................................................................. 101 5.4 RESEARCH APPROACH ..................................................................................... 105 5.5 METHODOLOGICAL CHOICE ........................................................................... 107 5.6 RESEARCH STRATEGY ...................................................................................... 113 5.7 TIME HORIZON .................................................................................................... 114 5.8 RESEARCH TECHNIQUES AND PROCEDURES ............................................. 115 5.8.1 Sampling ....................................................................................................... 115 5.8.2 Data Collection ............................................................................................. 120 5.8.3 Data Analysis ................................................................................................ 133 5.9 CONCLUSION ....................................................................................................... 136 CHAPTER 6: DISCUSSION AND INTERPRETATION OF RESULTS ............ 137 6.1 INTRODUCTION .................................................................................................. 137 6.2 DESCRIPTIVE STATISTICS ................................................................................ 137 6.2.1 Green Appliance Product Awareness ........................................................... 138 6.2.2 Green Appliance Product Purchase/Usage ................................................... 138 6.2.3 Participants’ Profile ...................................................................................... 139 6.2.4 Measurement Scales ..................................................................................... 141 6.2.4.1 Functional Value ................................................................................ 141 6.2.4.2 Social Value ........................................................................................ 142 6.2.4.3 Emotional Value ................................................................................. 142 6.2.4.4 Conditional Value ............................................................................... 143 6.2.4.5 Epistemic Value .................................................................................. 144 6.2.4.6 Relative Advantage ............................................................................. 145 ix 6.2.4.7 Compatibility ...................................................................................... 145 6.2.4.8 Complexity .......................................................................................... 146 6.2.4.9 Observability ....................................................................................... 147 6.2.4.10 Attitudes ............................................................................................ 147 6.2.4.11 Perceived Behavioural Control ........................................................ 148 6.2.4.12 Behavioural Intention ....................................................................... 149 6.2.4.13 Adoption Behaviour .......................................................................... 150 6.3 MODEL ASSESSMENT ........................................................................................ 151 6.3.1 Measurement Model ..................................................................................... 151 6.3.2 Structural Model Analysis ............................................................................ 161 6.3.2.1 Test of Collinearity ........................................................................... 162 6.3.2.2 Structural Model Results and Observation of Path Coefficients ..... 163 6.3.2.3 Discussions on Results from Hypotheses Testing............................. 165 6.3.2.4 Evaluating the Explanatory Power of the Predictors with the Coefficient of Determination ( 𝑅𝑅2) .............................................................. 171 6.3.2.5 Evaluation of the Effect Size (ƒ²) ...................................................... 173 6.3.2.6 Assessment of Predictive Relevance (𝑄𝑄2) ........................................ 174 6.4 CONCLUSION ....................................................................................................... 175 CHAPTER 7: CONCLUSION AND RECOMMENDATIONS ............................. 177 7.1 INTRODUCTION .................................................................................................. 177 7.2 OVERVIEW AND RESEARCH OBJECTIVES OF THE STUDY ...................... 177 7.2.1 Overview of the Study .................................................................................. 177 7.2.2 Research Objectives ...................................................................................... 177 7.2.3 Summary of Chapters ................................................................................... 178 7.3 KEY EMPIRICAL FINDINGS AND CONCLUSIONS OF THE STUDY .......... 179 7.3.1 Extent to which Consumer-related Factors Impact on Consumer Attitudes 180 7.3.2 Extent to which Product-related Factors Impact on Attitudes Towards Green Products ................................................................................................................. 183 7.3.3 Extent to which Green Product Attitude, Social Value (Subjective Norm) and Perceived Behavioural Control Affect the Intention to Buy Green Products . 185 x 7.3.4 Relationship Between the Intention to Buy Green Products and Actual Behaviour Towards Green Products ...................................................................... 186 7.3.5 Observations on Testing the Integrated Conceptual Model for Understanding Consumer-and Product-related Factors that Impact on Green Product Purchase Behaviour ............................................................................................................... 187 7.4 CONTRIBUTION OF THE STUDY ..................................................................... 192 7.4.1 Theoretical Contribution ............................................................................... 192 7.4.2 Practical Contribution ................................................................................... 193 7.5 LIMITATIONS OF THE STUDY AND RECOMMENDATIONS FOR FUTURE RESEARCH .................................................................................................................. 195 7.6 CONCLUDING REMARKS .................................................................................. 196 REFERENCES ............................................................................................................ 198 ANNEXURE A: QUESTIONNAIRE COVER LETTER ........................................... 254 ANNEXURE B: RESEARCH QUESTIONNAIRE .................................................... 255 ANNEXURE C: ETHICAL CLEARANCE LETTER ................................................ 262 xi LIST OF FIGURES Figure 2.1: The pyramid of green marketing .................................................................. 33 Figure 2.2: Population size of the middle class in South Africa ..................................... 44 Figure 2.3: Electrical home appliances owned by the black middle class ...................... 48 Figure 4.1: The Theory of Reasoned Action .................................................................. 71 Figure 4.2: The Theory of Planned Behaviour ............................................................... 74 Figure 4.3: The Theory of Consumption Values ............................................................ 76 Figure 4.4: Stages in the innovation diffusion process ................................................... 81 Figure 4.5: Adopter categories ........................................................................................ 82 Figure 4.6: Conceptual research model .......................................................................... 94 Figure 5.1: Research onion ........................................................................................... 101 Figure 6.1: Structural model with path coefficients and path significance................... 165 Figure 6.2: Structural model with path coefficients and 𝑅𝑅2 ......................................... 172 Figure 7.1: The structural model with path coefficients and 𝑅𝑅2 estimates ................... 188 xii List of Tables Table 2.1: Environmental innovation classification system ........................................... 24 Table 2.2: Studies on the characteristics of green products ............................................ 25 Table 2.3: Internal and external barriers to supply and marketing of green products .... 36 Table 5.1: Positivist and interpretivist research philosophies ....................................... 104 Table 5.2: Deduction, induction and abduction research approaches ........................... 106 Table 5.3: Summary of quantitative and qualitative research methodologies .............. 112 Table 5.4: Summary of characteristics of the major research strategies ....................... 114 Table 5.5: Sampling designs ......................................................................................... 115 Table 5.6: Types of survey research methods ............................................................... 121 Table 5.7: Scales of measurement ................................................................................ 124 Table 5.8: Operational definitions of the study constructs ........................................... 125 Table 6.1: Awareness of the green products ................................................................. 138 Table 6.2: Purchased/used green appliances ................................................................. 138 Table 6.3: Number of green appliances used ................................................................ 139 Table 6.4: Demographic information ............................................................................ 140 Table 6.5: Functional value ........................................................................................... 141 Table 6.6: Social value .................................................................................................. 142 Table 6.7: Emotional value ........................................................................................... 143 Table 6.8: Conditional value ......................................................................................... 143 Table 6.9: Epistemic value ............................................................................................ 144 Table 6.10: Relative advantage ..................................................................................... 145 Table 6.11: Compatibility ............................................................................................. 145 xiii Table 6.12: Complexity ................................................................................................ 146 Table 6.13: Observability.............................................................................................. 147 Table 6.14: Attitudes ..................................................................................................... 148 Table 6.15: Perceived behavioural control ................................................................... 149 Table 6.16: Behavioural intention................................................................................. 150 Table 6.17: Adoption behaviour ................................................................................... 152 Table 6.18: CFA analysis criteria for the measurement model .................................... 153 Table 6.19: Factor loading for initial instrument .......................................................... 154 Table 6.20: Internal consistency reliability ................................................................... 157 Table 6.21: Average variance extracted........................................................................ 158 Table 6.22: Discriminant validity ................................................................................. 159 Table 6.23: Factor loadings and cross loadings ............................................................ 159 Table 6.24: Criteria for assessing structural model ...................................................... 162 Table 6.25: Collinearity statistic – Variance inflation factors (VIF) ............................ 163 Table 6.26: Summary of structural model analysis ...................................................... 165 Table 6.27: Effect size (𝑓𝑓2) .......................................................................................... 173 Table 6.28: Predictive relevance (𝑄𝑄2 value) ................................................................. 175 Table 7.1: Summary of studies that have used either one or two theories (TPB, TCV, and DOI) to understand green consumer behaviour ..................................................... 190 1 CHAPTER 1: INTRODUCTION AND BACKGROUND 1.1 INTRODUCTION There has been an increased global interest in green products, which have shown a remarkable growth in sales over the last ten to twenty years (Koller, Floh and Zauner, 2011:1163). The reasons for this growth, according to Vermeulen and Bienabe (2010:20), revolve around environmental, health and social concerns which are influencing consumer demand. Consumers are thus becoming more concerned about the products they use in their households and the food they eat, not only in terms of intrinsic factors (e.g. chemical residues, environmental contamination and nutritional content) that affect health, but also with regard to broader environmental issues such as the impact on the environment (biodiversity and climate change) and the welfare of animals. The term ‘green products’ is defined as products that do not pollute the earth or deplete natural resources and can be recycled or conserved (Shamdasani, Chon-Lin and Richmond, 1993:489; Mostafa, 2007:220). According to Tan (2011a:1766), green products address environmental issues by having characteristics of recyclability, reusability, refillability, long life, degradability or decomposability, high quality in terms of their green performance, energy saving and being made of recycled materials. Green products do not only help the environment; they are also a source of competitive advantage, profitability and a means to tap into the large and growing segment of health and environmentally conscious consumers. To reap the rewards of green products, firms and nations need to be committed to the innovation and production of green products (Olson, 2013:792). In South Africa (SA), for example, significant progress has been made with regards to environmental management in the last decade by implementing laws and strategies that focus on sustainable development and green issues (United Nations Environment Program, 2011:22). A number of South African businesses that in the past assumed that incorporating ‘green’ into their business strategy would cost money, are now realising that ignoring negative impacts on the environment will, in fact, be costly in the future (Van der Zee, 2008:6). 2 Green product innovation, production and expected market and financial performance depends on assessing and understanding the market demand for the products (Lin, Tan and Geng, 2013:101) and the factors which drive the adoption of the green product innovation (Olson, 2013:791). This is particularly so because the demand for conventional products still outstrips the demand for green products (Ha and Janda, 2012:462; Kilbourne and Pickett, 2008:887; Olson, 2013:791). According to Gleim, Smith, Andrews and Cronin (2013:46), the estimated market share for green products in the year of 2013, was less than 4% worldwide. These figures confirm the findings of Godelink (2012:1) who maintains that although consumer spending has generally increased for conventional products, the ‘green glass ceiling’ is still unbreakable and its market share is still very small. There is therefore a clear need to study the factors that either hinder or drive the purchase and consumption of green products, especially in South Africa, where very few studies have been conducted on consumption drivers of green products or on environmentally and ethically friendly products. Studies in South Africa have focused on understanding the ethical consumer (Hughes, McEwan and Bek, 2015:148), green habits of generation Y consumers (Anvar and Venter, 2014:183), consumer attitudes towards green cosmetics (Beneke, Frey, Deuchar, Jacobs and Macready, 2010:477), consumer perceptions of environmentally friendly packaging (Scott and Vigar- Ellis, 2014:642) and consumer behaviour in relation to green wines (Kruger and Saayman, 2013:63). These studies chiefly assessed awareness, perceptions, habits and knowledge of green products, however, the way in which these factors determine consumers’ willingness and actual purchase behaviour is yet to be investigated. Moreover, Lin et al. (2013:105) suggest that in today’s highly competitive market, green product marketers and innovators need to understand the value that consumers derive from their products to gain competitive advantage and sustain market demand. While Sheth, Newman and Gross’s (1991) Theory of Consumption Values (TCV) identifies the types of value consumers may gain from different product categories, Thøgersen, De Barcellos, Perin and Zhou (2015:411) suggest that for green products in emerging markets such as China and Brazil, the value consumers expect to derive from green products will affect their attitudes towards those products. According to the Diffusion of Innovation theory (DOI), it is the product-related factors which drive consumers’ attitude towards the products. These attitudes, according to the Theory of Planned Behaviour (TPB), can lead to purchase intention, which is a driver of purchase behaviour. 3 This study integrated the TCV, DOI and TPB to examine the extent to which consumer- and product-related factors impact on green product purchase behaviour in South Africa. Considering that high household electricity consumption is a growing problem in South Africa, and 80% of middle class consumers use electric cookers and appliances for household cooking, cooling, heating and entertainment (Department of Energy, 2013:13), this study will focus on green household electrical products such as fridges, washing machines, dryers, heaters, stoves, irons and light bulbs. The South African black middle class was the subject of study as this consumer market segment is said to be growing rapidly in size, influence and income. It also has the largest spending power compared to other South African middle class segments and, most importantly, its expenditure on houses, cars and household technological and electrical products is increasing rapidly (University of Cape Town’s Unilever Institute of Strategic Marketing, 2013:1). Considering the environmental and resource conservation benefits of consuming green products, this study will practically and theoretically contribute by revealing the consumer and product-related factors driving green products attitudes and ultimate purchase intention and behaviour of a large and lucrative market segment, such as the black middle class in South Africa. With this knowledge, marketers would be able to not only promote the appropriate benefits of purchasing green products over conventional products, but also increase their market demand for green products. 1.2 PROBLEM STATEMENT Numerous studies have been conducted on understanding green purchase behaviour. For example, in Asia researchers (Biswas and Roy, 2015; Kumar, 2012; Chen and Chang, 2012; Wu and Chen, 2014; Lin and Huang, 2012; Huang, Lin, Lai and Lin, 2014; Tan, 2011; Onwezen, Bartels and Antonides, 2014; Chairy, 2012; Zhao, Gao, Wu, Wang and Zhu, 2014) have examined the main determinants of green cosmetic product consumption, recycling, wind energy and influences of culture in green product purchases. In the United State of America, researchers (Matthes, Wonneberger and Schmuck, 2014; Barber, Bishop and Gruen, 2014; Elliott, 2013; Cho, Thyroff, Rapert, Park and Lee, 2013) have looked at the effects of advertising in increasing the purchase of green products, the role of price in green product purchase, consumer perceptions regarding hybrid vehicles and the role of culture in environmental behaviour. In Europe, studies (Vermeir and Verbeke, 2008; Gracia and 4 Magistris, 2007; Leire and Thidell, 2005; Grankvist and Biel, 2007; Kanchanapibul, Lacka, Wang and Chan, 2014; de Maya, Lopez-Lopez and Munuera, 2011; Paco, Alves, Shiel and Leal Filho, 2013) sought to understand perceptions of customers regarding green products, the influence of cultural values on green product purchase and the importance of eco-labelling in improving sales of green products. These studies have mainly and independently used either the Theory of Planned Behaviour (TPB) or the Theory of Consumption Values (TCV) to understand green products purchase behaviour. However, with the complexity of drivers of purchase behaviour, especially for non-conventional items such as green products, there is a need to integrate theories for a deeper understanding of the phenomenon. The Theory of Planned Behaviour (TPB) has been applied across a variety of pro- environmental behaviour studies, including understanding the key factors that influence the intention to consume green products (Gracia and Magistris, 2007; Wu and Chen, 2014; Kim and Han, 2010), recycling (Boldero, 1995; Taylor and Todd, 1995) and reducing energy consumption (Sparks, Hindis, Curnock and Pavey, 2014). Findings from these studies have supported the usefulness of the TPB constructs in predicting the intention to go green or to adopt green behaviour. However, the strength of the explanatory power of the TPB model has been widely debated (Eagly and Chaiken, 1993; Conner and Armitage, 1998) and it has been suggested that more constructs should be examined in the TPB to improve the variance explained of purchase intention or behaviour. Thøgersen et al. (2015:411) maintain that such constructs may be the different values consumers enjoy from green products. The Theory of Consumption Values (TCV) of Sheth et al. (1991) suggests the types of values which consumers may seek from green products. The TCV has been used extensively in the past in analysing consumer behaviour. In their attempt to understand green behaviour, Lin and Huang (2012:11) used the consumption values to determine the main influencing factor on choice behaviour. Haws, Winterich, Naylor (2014:336) and Suki and Suki (2015:269) examined the impact of green consumption values on protecting the environment. Goncalves, Lourenco, Silva (2016:1484) adopted the TCV in identifying the specific consumption values that can better predict green purchase behaviour. Biswas and Roy (2015:332) and Hur, Yoo and Chung (2012:688) also used the TCV in understanding environmental attitudes, consumer innovativeness in relation to sustainable green consumption. Thøgersen et al. (2015:411) suggest that in emerging countries, the purchase of green products is not only rooted in consumer value priorities, but also in how the values drive attitudes towards green products in 5 the first place. All these studies uncover the various values consumers would have to enjoy in order to develop positive attitudes, go green and act pro-environmentally. However, these researchers did not examine the product-related factors which may drive attitudes and help or hinder green product purchase. The DOI suggests such factors. The DOI has been widely used to identify aspects of an innovation (such as a product or idea), which helps or hampers the adoption of the innovation. The theory has also been used to predict whether or how a new invention will be successful (Zhu, Dong, Xu and Kraemer, 2006:602; Tan, Chong, Lin and Eze, 2008:227; MacVaugh and Schiavone, 2010:198). Even though the DOI is mainly used to determine the adoption of technological, business-to-business products, a study by Vaccaro (2011:328) demonstrates that the DOI can provide insights into understanding product-related factors driving the adoption of green consumer products. Considering, however, that the intention and actual adoption of behaviour may be first driven by a positive attitude towards the innovation, as the TPB suggests, it is important to integrate both the DOI and the TPB to better understand the process of green product adoption. In South Africa, studies are needed to understand green products purchase drivers to improve demand. Local research has chiefly focused on assessing the level of awareness, perceptions, interest and environmental concerns regarding eco-friendly products or green products with little insights into how these factors drive green product purchase behaviour. The present study will therefore integrate the TPB, TCV and DOI to provide a holistic look at the consumer and product-related factors driving green product purchase in South Africa. With the growing awareness and interest in green products and green marketing in SA (Dallas, 2015:14), this study will be of value to companies seeking to understand what consumers seek in green products and which aspects of those products will affect their attitudes and ultimate purchase and consumption. 1.3 OBJECTIVES OF THE STUDY 1.3.1 Primary Objective The primary objective of this study is to investigate the consumer and product-related factors impacting on green appliance attitudes and purchase behaviour among the black middle class in South Africa. 6 1.3.2 Secondary Objectives To achieve the primary objective, the following theoretical and empirical secondary goals have been formulated: 1.3.2.1 Theoretical Objectives  To assess the size of supply and demand of green products globally and in South Africa and the attractiveness of Black middle class as a market for green products  To review literature on the consumer-related factors that impact on consumers’ attitude towards the purchasing of green products.  To review literature on the various product-related factors that impact on consumers’ attitude towards the purchasing of green products.  To review literature on the extent to which consumers’ attitude towards green products can affect their intention and decision to purchase green products.  To appreciate theories and models explaining green products purchase, and to develop a conceptual model, which delineates how consumer and product-related factors affect attitude towards green products, purchase intention and actual behaviour. 1.3.2.2 Empirical Objectives  To examine the extent to which functional, emotional, social, conditional and epistemic values (i.e. consumer factors) impact on consumer attitude towards green products.  To examine the extent to which relative advantage, compatibility, complexity and observability (i.e. product-related factors) impact on attitude towards green products.  To determine the extent to which green product attitude, social value (subjective norm) and perceived behavioural control affect the intention to buy green products.  To examine the relationship between the intention to buy green products and the actual behaviour towards green products.  To empirically test the conceptual model developed to understand the consumer and product-related factors that impact on green product purchase behaviour. 7 1.4 RESEARCH QUESTIONS This study seeks to answer the following main question: → To what extent do the consumer- and product-related factors impact on attitudes towards green products and the ultimate intention and actual purchase behaviour of black middle class in South Africa? From this question, the following sub-questions will be addressed:  What are the impacts of functional, emotional, social, conditional and epistemic values (i.e. consumer factors) on attitudes towards green products among South African black middle class?  What are the impacts of relative advantage, compatibility, complexity and observability (i.e. product-related factors) on attitudes towards green products among South African black middle class?  What influence do attitudes towards green products, social value (subjective norm) and perceived behavioural control have on the intention to purchase green products among South African black middle class?  To what extent does the intention to buy green products affect the actual purchase among South African black middle class? 1.5 PRELIMINARY LITERATURE REVIEW This section reviews literature on green consumer behaviour. It then discusses a number of theories that are commonly used in understanding green purchase behaviour by consumers. The relevant literature will be drawn from textbooks, journal articles and conference papers. 1.5.1 Overview of Green Consumer Behaviour An understanding of green consumer behaviour is fundamental for environmental and business reasons. From an environmental standpoint, the reduction of negative effects of consumption is essential to fulfil some of the goals put forth by the international community (Organisation for Economic Co-operation and Development - OECD, 2009; United Nations Environment Programme - UNEP, 2007; 2010). From a business and marketing perspective, the development of less ecologically harmful products is of little use without consumers adopting these greener 8 technologies and lifestyles. Researchers have also argued that the ‘right’ purchasing decision has the potential to reduce, and even eliminate, environmental harm in the later stages of the consumption cycle (Cheah and Phau, 2011:503; Vergragt, Akenji and Dewick, 2014:2). However, the consumer behaviour of reducing negative environmental impact (for example, energy conservation and recycling) has received considerably more attention than the consumer behaviour of purchasing green products. The failure to purchase green products, Tseng and Hung (2013:182) suggest, could be attributed to the unavailability of green products. However, even when green products are available, it is necessary to understand the factors helping or hindering green consumer behaviour (Follows and Jobber, 2000:745; Cheah and Phau, 2011:452). Consequently, theories and models have been proposed to understand the drivers of green consumer behaviour. 1.5.2 Diffusion of Innovation Theory Diffusion of innovation is defined as a process by which an idea, practice or object that is perceived as new catches on among mainstream consumers. It is described as a process by which consumers gain knowledge about the innovation, are persuaded or form attitudes towards the innovation, make a decision to adopt or reject the innovation, use the innovation, and evaluate the results (Rogers, 2003:229). Rogers (2010:62) posits that every time that an innovation emerges, the act of diffusion occurs, involving a process of communicating or disseminating the information and opinions on that innovation to potential users through a communication channel. Based on this information, the potential adopter then decides to adopt (or reject), and use the innovation to confirm his/her decision. Rogers (2010:10) refers to this process as the innovation decision process and conceptualises five logical steps of this process as depicted in Figure 2.1 (see Chapter 2). The steps in the process are: knowledge, persuasion, decision, implementation and confirmation. Rogers (2003:232) further states the rate of adoption id affected by factors such as type of innovation decision, communication channels involved, nature of the social system in which the innovation is placed and the extent of the change agents’ promotional efforts. Rogers (2003:232) argues that perceived innovation characteristics are an important explanation of the rate of adoption and that most of the variance in adoption rates of (49 to 67%) is explained by five characteristics. The five innovation characteristics are as follows: 9 Relative advantage – This is described by Rogers (1995:11) as the degree to which an innovation is perceived to be a better idea than the idea it supersedes. For example, a new product that is technologically more advanced and may reduce production costs, may be evaluated as having a relative advantage. Compatibility – This refers to the degree to which an innovation is perceived as being consistent with existing values, past experiences and needs of potential adopters. Complexity – This refers to the degree to which an innovation is perceived as relatively difficult to understand and use. The more difficult a product is to understand and use, the less likely consumers are to adopt the product. Trialability – This refers to the degree to which an innovation may be experimented with, on a limited basis. According to Rogers (1995:12), when users can try an idea in small doses or ease into an innovation gradually, they are more likely to adopt the innovation. This innovation characteristic will be excluded from this study. Previous research on household/domestic solar power systems has excluded trialability, as these products are not considered to be trialable due to their nature (Faiers and Neame, 2006; Tapaninen, Seppanen, and Makinen, 2009a:102) Observability – This refers to the extent to which the outcomes of an innovation are noticeable to others. The easier it is for potential users to see the results of an innovation, the more likely they are to adopt it. Although the DOI theory has been widely accepted as a useful theory for studying the adoption of technological innovations, it can also be used to understand the adoption of green products. For example, the DOI has been used to explore factors driving the decision to adopt or reject a green product innovation or stimulate the diffusion of green products (Oliver, 2007:24; Janssen and Jager, 2002:288), consumer purchase motivations for hybrid cars (Ozaki and Sevastyanova, 2011:2218) and modelling of eco-innovation diffusion (Kijek, 2015:67). Another useful theory, which can be used for the understanding of green product purchase behaviour is the TPB. 10 1.5.3 Theory of Planned Behaviour The Theory of Planned Behaviour (TPB), as proposed by Ajzen in 1991, is derived from the Theory of Reasoned Action developed by Fishbein and Ajzen (1975). According to the TPB, an individual’s specific behaviour is determined by his/her behavioural intention to perform the behaviour. This behavioural intention is in turn determined by three factors related to the behaviour: the person’s attitude, subjective norms and perceived behavioural control. Attitude reflects the degree to which carrying out the behaviour is valued positively or negatively by the individual, subjective norm reflects the perceived social pressure to engage with the behaviour and perceived behavioural control reflects the individual’s appraisal of his/her ability to carry out the behaviour. In summary, if individuals feel favourable towards a particular behaviour, believe that important people in their lives would approve of them carrying out the behaviour and are confident about their ability to undertake the behaviour, then they are more likely to have a positive intention to engage in the behaviour. This positive intention is in turn associated with a greater likelihood that the behaviour will be enacted. It is also worth noting that in some situations perceived behavioural control is believed to exert a direct effect on behaviour unmediated by intention (Ajzen, 1991:198). According to Han, Hsu, Lee and Sheu (2011:351), the TPB allows researchers to examine the influence of personal determinants, social surroundings as well as non-volition determinants on intention. The TPB model optimises the potential relationship between intention and its determinants by measuring each construct at equivalent levels of specificity. The TPB has been used to understand organic food choice (Paul and Rana, 2012:415; Davis, O’Callaghan and Knox, 2009:138) and green purchase intentions (Chen and Tung, 2014: 222; Barber, 2010:424). Although the TPB gained empirical support from previous pro-environmental behaviour studies (Dowd and Burke 2013:137; De Leeuw, Valois, Ajzen and Schmidt, 2015:128), it has inherent limitations (Sniehotta, Presseau and Araujo-Soares, 2014:3). The TPB emphasises proximity between intention and behaviour, which renders the operationalisation of the TPB problematic. These limitations notwithstanding, Ajzen (1991:199) maintains that TPB is open to the inclusion of additional constructs that have the ability to demonstrate a significant variance in intention or behaviour. 11 1.5.4 Theory of Consumption Values (TCV) The Theory of Consumption Values proposed by Sheth, Newman and Gross (1991a, 1991b) suggests that consumers attach different values to product groups and that these in turn will affect motivations to purchase. Sheth et al. (1991) suggest that consumers’ choice behaviour of products or services is influenced by five expected values or benefits, namely, functional, social, conditional, emotional and epistemic values. The five values are explained as follows: Functional value - also referred to as ‘value for money’, this is measured by consumers’ perception about the product performance with respect to price. It is assessed as the primary driver of consumer choice behaviour with regard to green product purchase (Sheth et al., 1991:172; Biswas and Roy, 2015:464). Social value – measures perceived utility derived from consumers’ association with social groups. Social value for green products can be defined as the perceived net utility derived from green product consumption based on the perception of social pressure or prestige gain through engagement with environmental issues (Sheth et al., 1991:172; Biswas and Roy, 2015:464). This represents the subjective norm in the TPB. Conditional value – denotes the perceived net utility derived in a specific situation or circumstances faced by the decision maker (Sheth et al., 1991:172; Gadenne, Sharma, Kerr and Smith, 2011:7686). Emotional value – refers to the perceived utility derived from an alternative capacity to arouse feelings or affective state. (Sheth et al., 191:173). Epistemic value – refers to the perceived utility resulting from a product or service that stimulates the desire for knowledge and offers novelty (Sheth et al., 1991:173; Biswas and Roy, 2015:464). 12 Sheth et al. (1991b:6) posit that by analysing the values which predominate among a given population regarding a product category, it is possible to discriminate between various offers according to particular value weightings. The TCV applies only in cases of individual, voluntary and rational or systematic decision situations, and depends on three axioms: i. Market choice is a function of multiple values; ii. These values make differential (and incremental) contributions to choice; and iii. The values are independent (Sheth et al., 1991b:7). The TCV has been applied by Lin and Huang (2012) in determining the drivers of consumer choice behaviour for green products. Biswas and Roy (2015) also used TCV to examine the consumption values impacting on sustainable consumption behaviour across different consumer segments with preferential green choice. While Lin and Huang (2012) found all consumption values have a positive relationship with the choice behaviour of green products, Biswas and Roy (2015) revealed that social value was the dominant factor in influencing sustainable consumption behaviour adoption among the green consumers. Whether these values will affect South African consumers’ attitudes toward green product purchase is one of the aims of this study. 1.6 CONCEPTUAL MODEL AND HYPOTHESIS DEVELOPMENT This study integrates the components of the DOI, TPB and TCV to develop the conceptual model and hypotheses. This conceptual model proposes that the consumer-related factors (functional, social, emotional, conditional and epistemic values) and product-related factors (relative advantage, compatibility, complexity and observability) are likely to positively influence green purchase behaviour attitudes, which may in turn positively lead to intention to buy and the actual buying of green products. Social value (subjective norm) is likely to positively influence behavioural intention. Perceived behavioural control is expected to influence the intention to buy and the actual purchase of green products. The Theory of Consumption values developed by Sheth et al. (1991:159) focuses on values that explain “why consumers choose to buy or not to buy (or use or not use) a specific product, why consumers choose one product type over another”. The Diffusion of Innovation theory of 13 Rogers (2003:273) is mostly used to study adoption patterns, clarify the mechanism and assist in predicting whether and how a new invention or idea will be successful. The Theory of Planned Behaviour, which was first put forward by Ajzen (1991:197), is one of the classic theoretical models for studying individual behaviour. The theory suggests that an individual’s behavioural intention is affected by behavioural attitudes, subjective norms and perceived behavioural control. A number of studies that have examined the key determinants of green products have integrated the TCV and TPB (de Maya et al., 2011:1773, Lin and Huang, 2012:15; Papista and Krystallis, 2013:83; Jiang and Kim, 2014:308; Biswas and Roy, 2015:336). All of the results from these studies show that the integrated models provided a good relationship of green product purchase attitudes and behavioural intention. In the study by Biswas and Roy (2015:335), their results for testing the integrated model explained 51% variance in attitudes and 53% variance in behavioural intention. The integrated model of DOI and TPB has been used in numerous studies seeking to understand the important attributes that consumers or potential consumers will consider when buying a new cell phone, hybrid vehicle, alternative fuel vehicle or solar system (Jansson, 2011:196; Ozaki and Sevastyanova, 2011:2218; Karakaya, Hidalgo and Nuur, 2014:393). All these studies showed a good relationship between variables and attitudes towards these above- mentioned products. In the study (Karakaya et al., 2014:393) on hybrid vehicles (also referred to as eco-innovation in the study), the integrated model was found to explain 52% and 49% of the variance in attitudes towards hybrid vehicles and behavioural intention respectively. It is therefore, the researchers’ opinion that the proposed integrated model from three theories is likely to provide a higher explanatory power for attitudes towards green product purchase, intention and actual behaviour than when these theories are used separately or paired. The next section introduces the research methods used to test the conceptual model. 14 1.7 OVERVIEW OF RESEARCH METHODOLOGY The research methodology outlines the procedures that are followed by a researcher when conducting the study (Saunders, Lewis and Thornhill, 2012: 118). This study used the following research methodology described below. 1.7.1 Research Philosophy A research philosophy refers to a set of assumptions and beliefs that guide an empirical study (Jonker and Pennink, 2010:29). Saunders et al. (2012:119) propose three research philosophies − positivism, realism and interpretivism. This study follows a positivistic philosophy. The positivistic philosophy assumes that objective knowledge is acquired through the use of systematic, scientific methods of inquiry (Bryman and Bell, 2011:20). Using theories and models, positivists formulate hypotheses which need to be tested (Saunders et al., 2012:120; Sekaran and Bougie, 2013). Consistent with the positivistic philosophy, this study proposed a set of hypotheses that were verified by collecting quantitative data that were subjected to statistical analysis. 1.7.2 Research Approach From the inductive and deductive research approaches described by Saunders et al. (2012:121), this study subscribed to the deductive approach. While the inductive approach is aimed at developing a theory or a model, the deductive is used to test theories or models. The deductive approach draws inferences from analysed data with the objective of generalising research findings to a broader population (Saunders et al., 2012:121). The deductive approach is consistent with the objectives of this study, which tests relationships between variables as hypothesised in the conceptual model. 1.7.3 Research Strategies Saunders et al. (2012:122) define a research strategy as the overall plan that is followed by a researcher to gather the primary data and answer the research questions. A survey is an appropriate strategy for this study. Survey research involves acquiring information about one or more groups of people about their characteristics, opinions and attitudes, or previous experiences, by asking them questions and tabulating their answers. The ultimate goal is to learn about a large population by surveying a sample of that population (Leedy and Ormrod, 2013:195). According to Neuman (2011:83), surveys give the researcher a picture of what many 15 people think or report doing and also allow the researcher to explore the characteristics, attitudes and behaviours of the targeted respondents through interviews or a self-administered research instrument. 1.7.4 Data Collection Methods or Designs The primary data collection was accomplished through a survey using structured questionnaires. It was a face-to-face survey, which produces a higher response rate and is known to generate large amounts of data (Babin and Zikmund, 2016:303). The use of questionnaires during a survey yields essential information which can be generalised to the target population (Saunders et al., 2012:122). 1.7.5 Sampling Design The target population for this study was the South African black middle class living in Gauteng Province, who have the income and capability of purchasing green appliances. A study conducted by the University of Cape Town’s (UCT) Unilever Institute of Strategic Marketing (2013) posits that more than 4,2 million people in South Africa fall within the category of black, middle class, salaried and well-educated people. Gauteng Province is considered to be the ideal setting for this study as more than 66% of the black middle class resides in Gauteng, according to the UCT findings. The UCT findings also indicate that the spending power of this market has increased substantially, from R1,7 billion (approximately, $123 million US dollars) annually in 2004 to over R400 billion (approximately, $29 billion US dollars) annually in 2013. The findings further state that more people are entering the housing market for the first time, spurring construction and resulting in increased demand for household products. There are two types of sampling techniques, namely, probability and non-probability sampling (Burns and Bush, 2014:262; Saunders et al., 2012:125). Probability sampling is “a sampling technique in which each population element has a known, non-zero chance of being included in the sample” (Iacobucci and Churchill, 2010:285). The non-probability sampling technique, on the other hand, relies on the researcher’s judgment to select a suitable population element to form part of the sample (Zikmund and Babin, 2013:305). Probability sampling is constituted by simple random sampling, stratified sampling, systematic sampling and cluster sampling. Non-probability sampling is subdivided into convenience sampling, quota sampling, judgement sampling and snowball sampling (Saunders et al., 2012:125). For the purposes of this study, a 16 probability sampling technique using simple random sampling was used. This was possible, because the research company that was tasked with the field work had an existing database for the South African middle class from which they could randomly select respondents to participate in the study. Simple random sampling ensures a high degree of representativeness, as it guarantees that every sample of a given size, as well as every individual in the target population, has an equal chance of being selected (Saunders et al., 2012:125). 1.7.6 Determining the Sample Size The sample size for the study will be determined by using two approaches − the historical evidence method and the requirements of the data analysis method. The historical evidence method determines sample size by benchmarking against previous related studies (Zikmund and Babin, 2010:265). Based on previous similar studies conducted by Arvola, Vassallo, Dean, Lampila, Saba, Lahteenmaki and Shepherd (2008:446), Orten and Atik (2009:822), Thøgersen (2011:1062), Peloza, White and Shang (2013:106), White and Simpson (2013:81), Rajalakshmi (2014:143), Cheung, Lam and Lau (2015: 233), it was found that researchers used sample sizes of between 200 – 400. Based on the requirements of the data analysis methods, as a rule of thumb for robust partial least squares structural equation modelling (PLS-SEM) estimations, it has been recommended that the sample size be equal or larger to ten times the number of structural paths directed at a particular construct in the structural model (Barclay, Higgins and Thompson, 1995: 306). Kline (2011:112) further states that when using confirmatory factor analysis, having fewer than 300 respondents is considered weak, 300 respondents is considered to be good, and 400 to 500 respondents is very good. With this study having thirteen constructs, the sample size of 500 respondents was deemed acceptable for the confirmatory factor analysis and partial least square structural equation modelling that was used to analyse data of this study. 1.7.7 Sampling Procedure To contact the targeted sample, the researcher used the services of Ramsay Research Media – a research company that has offices in both Cape Town and Johannesburg. For more than ten years, Ramsay Research Media has been collecting large amounts of data for the UCT Unilever Institute of Strategic Marketing for numerous studies involving the black middle class of South Africa. Ramsay Research Media therefore has extensive experience in working with the South African black middle class and they could be contracted to collect reliable data for this study. 17 1.7.8 Questionnaire Design Data was collected by means of a self-administered, structured questionnaire. The questionnaire consisted of five sections. A cover letter introduced the purpose of the study and advised the participants of their anonymity, the confidentiality of their response and their right to withdraw from the study at any point in time, should they wish to do so. Section A contained questions that sought biographic information of the respondents using a multi-choice and dichotomous questions. Section B covered questions on consumer-related factors expected to have an impact on attitudes towards green products, such as functional, social, emotional, conditional and epistemic values. Section C covered questions on product-related factors expected to have an impact on attitude towards green products such as relative advantage, compatibility, complexity and observability. Section D questions were designed to measure respondents’ perceived behavioural control, attitudes, intention and actual behaviour of green product purchase. A filter question was used to separate current users and non-users of green products. All measurement items in Sections B, C, and D used a seven-point Likert scale anchored by ‘strongly agree’ (7) and ‘strongly disagree’ (1). All constructs in Sections B, C and D were operationalised using validated scales from previous similar studies (Beck and Ajzen, 1991; Kim and Choi, 2005; Mostafa, 2009; Han, Hsu and Sheu, 2010; Chen and Peng, 2012; Chen and Chang, 2012) in order to enhance reliability and validity. 1.7.9 Data Analysis Data analysis is defined as the process of bringing order, structure and meaning to the mass of collected data (De Vos, 2011:335). Descriptive statistics, reliability and validity tests, measurement model and PLS-SEM were used to analyse the data. The descriptive statistics provided the mean, standard deviation and percentages of respondents' responses. They were also used to assess the profiles of the respondents. A confirmatory factor analysis (CFA) using the partial least squares structural equation modelling technique, with the aid of the SmartPLS version 3.2.7 software, was conducted to evaluate the measurement model. A number of criteria were used to assess the measurement model; these included Factor Loading, Internal Consistency Reliability (Cronbach’s alpha and Composite Reliability), Convergent Validity and Discriminant Validity, as suggested by Hair, Hult, Ringle and Sarstedt, 2017). The same program was used to evaluate the structural equation model, which tested for Collinearity Assessment, Coefficients of Determination (𝑅𝑅2) 18 value, Predictive Relevance (𝑄𝑄2), identification of the size and significance of the path coefficient and effect sizes (𝑓𝑓2). The hypothesised relationships between the latent constructs in the structural model were also tested. 1.8 CONTRIBUTION OF THE STUDY This study seeks to make theoretical and managerial contributions. From a theoretical viewpoint, the following contributions will be made:  The body of knowledge on understanding the drivers of attitudes, intention and behaviour towards green product purchase and consumption will be broadened by conducting the study in a multicultural and socio-demographically diverse country like South Africa (SA).  This study will not only fill a void of academic information on drivers of green product consumption in South Africa, but will do so among the black middle class, who are a large, lucrative and fast-growing market segment in SA.  Another theoretical contribution of this study will be to test the explanatory power of integrating DOI, TPB and TCV and in an emerging consumer market. From a managerial viewpoint, the contributions will be the following:  Previous research studies have estimated market share for green products to be growing at a very slow pace despite increased awareness of green products. By understanding both the consumer and product-related factors helping and hindering green product purchase behaviour in SA, marketers will ultimately be better equipped with information on the benefits middle class consumers seek from green products and on how they can best position green products for increased adoption, purchase and consumption. This will not only increase sales and profits for the marketers from a large and lucrative market segment like the black middle class in SA, but will be good for the health of the consumer and the general environment. Other global marketers, consumer and environmental interest groups can use the findings for better targeting and positioning for green appliances. These can increase general demand for green appliances. 19 1.9 ETHICAL CONSIDERATIONS Ethical issues become crucial whenever human subjects are the focus of investigation. It is expected that the researcher always searches for truth, however, this should not be at the expense of others (Babbie and Mouton, 2008:521; Saunders et al., 2012:226). Babin and Zikmund (2016:89) suggests four ethical guidelines that should be adhered to by the researcher when dealing with respondents. These guidelines are listed as follows:  Respondents’ right to privacy should be preserved;  Respondents should be assured of confidentiality;  Respondents should not be deceived regarding the purpose of the study; and  Respondents’ safety should not be compromised. These guidelines were applied in this study. Furthermore, the questionnaire included a covering letter (Annexure A) which explained the purpose of the study and assured the respondents of their anonymity and confidentiality. The respondents were assured that the collected data would to be used for research purposes only. To ensure anonymity and confidentiality, no names were mentioned in the questionnaires (Babin and Zikmund, 2016:90). The findings of the study were reported in an aggregated format and no personal details of the participants were included in the presentation of results. In addition to the above, ethical clearance for the study was obtained from the University of the Witwatersrand internal ethics review committee, and a letter of approval authorising the researcher to proceed with the study was received (Annexure C). All data collected from respondents was stored and disposed of in accordance with the rules and regulations provided by the Ethical Review Committee. 1.10 OUTLINE OF THE STUDY This thesis comprises seven chapters which are outlined below. Chapter 1 – This chapter introduces the research and states the research problem and objectives. It also introduces the theories and models used to develop the conceptual model and hypotheses and describes the research methods which were used to test the model. 20 Chapter 2 – This chapter presents an overview of green product marketing, supply and demand. Green conscious consumers and the barriers and opportunities of green product consumption are also discussed. The chapter also discusses why understanding green consumption among the SA middle class is important for this study. Chapter 3 – This chapter reviews existing literature on drivers that have been found to influence green purchase behaviour. Chapter 4 – The theoretical models and theories that inform the drivers of green consumption are examined in this chapter. The conceptual model and hypotheses are also formulated in this chapter. Chapter 5 – This chapter outlines the research methodology of the study. Specifically, it discusses the research philosophy, research design, sampling method and other methods used to obtain the primary data, which was used for the empirical testing of the conceptual model. The chapter also describes the statistical techniques used to analyse the data and test the relationships of the conceptual model. Chapter 6 – The results of the statistical data analyses are presented and the key findings are highlighted in this chapter. Chapter 7 – Conclusions and key findings are discussed. Theoretical and practical contributions are presented, as well as limitations of the study and suggestions for future research of the study. 21 CHAPTER 2: GREEN PRODUCT SUPPLY AND DEMAND GLOBALLY AND IN SOUTH AFRICA: EXAMINING THE BLACK MIDDLE CLASS AS A MARKET SEGMENT 2.1 INTRODUCTION The risks and impact of climate change will increase significantly over the next decade (UNEP 2014a), with annual climate change adaptation costs projected to reach US$300 billion by 2050 (UNEP, 2015). Energy consumption, particularly in relation to carbon dioxide emissions, is the major contributor to rising global temperatures and many associated environmental problems (Gardner and Stern, 2002). The promotion and adoption of lower energy consuming lifestyles is therefore imperative for environmental sustainability (Bhate, 2005; Pothitou, Kolios, Varga and Gu, 2016; UNEP, 2015). This is particularly needed in emerging markets such as South Africa, where there are faster electricity demand growth rates (UNEP, 2014b). More so, in emerging markets, the economic aspirations of billions of inhabitants are often facilitated by “unprecedented levels of energy consumption” (Smil, 2013:145). Creating awareness of sustainability concerns and supplying green products to promote the demand and use of green products in emerging markets will therefore be important. Growing global public awareness of sustainability issues has been widely reported in documents about climate change (e.g. IPCC, 2014; UNEP, 2015) and numerous global initiatives such as the Efficient Appliances and Equipment Partnership (UNEP, 2014b) seek to accelerate the transition to more sustainable consumption behaviour. However, educating consumers about sustainability is challenging, not least because awareness and concerns about environmental issues frequently fail to translate into sustainable behaviours (Figueredo and Tsarenko, 2013). Nonetheless, according to Squires, Juric and Cornwell (2001), and depending on the level of economic development of a country, the demand for organic and green products is increasing. The question, however, is whether the supply of these products is matching the demand, and which consumer segment is the most likely to drive the demand of green products. 22 This chapter discusses the demand and supply of green products in South Africa and globally. It starts by exploring the concept of green products and green marketing, green product characteristics and how using green products can benefit the consumer and the community at large. It then proceeds to explore the supply of green products, barriers to supply and the size of the green market in South Africa and globally. The discussion moves to exploring the demand for green products and identifying the profiles of green product consumers. The chapter ends with a discussion of the South African black middle class as a potential market for green products, their consumption lifestyle and the attractiveness of the segment for green products purchase. 2.2 THE CONCEPT OF GREEN PRODUCTS AND GREEN MARKETING 2.2.1 What is a Green Product? The terms ‘green product’ or ‘environmentally-friendly products’ usually describe products that protect or enhance the natural environment, the conservation of energy and the reduction or elimination of toxic agents, pollution and waste (Olson, 2013:171). Olson (2013:172) further states that: …although no consumer product has a zero impact on the environment, in business, the terms ‘green product’ or ‘environmental product’ are used commonly to describe those that strive to protect or enhance the natural environment by conserving energy and/or resources and reducing or eliminating the use of toxic agents, pollution, and waste. The American Marketing Association (2015:1) defines green products as those which “use less resources, have lower impacts and risks to the environment and prevent waste generation already at the conception stage”. This definition emphasises the importance of designing green products, even from the conceptualisation phase. Peattie (2010:181) feels that a product is green “when its environmental and societal performance, in production, use and disposal is significantly improved in comparison to conventional or competitive products”. This means that the environment should be considered at the different life cycle phases of a product. Additionally, Tomasin et al. (2013:74) state that “green products are designed to prevent, limit, reduce and/or correct harmful environmental impacts on water, air and soil.” Reinhardt (1998:46) explains that environmental product differentiation entails “creating products that provide greater environmental benefits or that impose smaller environmental costs than similar products.” Dangelico and Pontrandolfo 23 (2010:1609) indicate that this definition illustrates how green products are able to generate a lower environmental impact and provide higher environmental benefits compared to conventional products. All these definitions stress the importance of considering environmental factors such as energy, resources, pollution, waste, water, air and soil at every stage of developing, manufacturing, using and disposing of products and services. Examples of some green products in the market are household appliances (microwaves, fridges, kettles, washing machines, hair dryers, heaters) that use less electricity and water; compact fluorescent lamps (CFLs) which use 75% less energy than incandescent bulbs (Energy Star, 2015); organic foods, which are grown and processed without antibiotics, pesticides or synthetic fertilisers (Environmental Protection Agency, 2015); hybrid gas electric vehicles such as Toyota Prius, which uses significantly less fuel than comparable conventional vehicles; and eco-friendly tissue papers and biodegradable detergents (European Commission, 2012b). Thus, in developing these green products, it is important to also create successful environmental strategies and to ensure that companies and economies become environmentally sustainable (Pujari, 2009:76). 2.2.2 Types and Characteristics of Green Products By studying literature on green product classification, it was discovered that different perspectives are used to classify green products. Dangelico and Pontrandolfo (2010:1609) quote Brezet’s (1997) studies that resulted in four different types of environmental innovations which vary in terms of eco-efficiency. As shown in Table 2.1 below, these are i) Product improvement, ii) Product re-design, iii) Function innovation and iv) System innovation. 24 Table 2.1: Environmental innovation classification system Type of Innovation Characteristics Product improvement  Product improvements from a preventative environmental impact perspective for existing products. Product and production technology are the same. Product re-design  The product concept remains almost intact. The product and its components are further developed or replaced, for example, by introducing non-toxic materials, improving distribution, recycling or energy efficiency. Function innovation  Not limited to existing product concepts. The innovation is related to how the function is achieved, for example changing paper-based information sources to e-mail. System innovation  The complete technological system (product, production chain, infrastructure and related institutions) is replaced by a new system. Source: Brezet (1997) These four types of innovation are all environmentally driven but have different characteristics. The first type involves the improvement of products from the perspective of pollution prevention and environmental care. Both the product and production technique will generally remain the same. With the second type of innovation which involves redesigning the product, the product concept remains the same but parts of the product are improved and replaced for the purpose of using non-toxic materials, recycling and disassembly, improved distribution and energy use reduction with respect to all components over the product life cycle. The starting point of the third type, functional innovation, is the function of the current product. In this case, the way in which the product fulfils the function is changed. The generation shift from physical products to dematerialised services belongs to this category. In the fourth type, the system innovation, the entire technological system (product, production chain and associated infrastructure and institutional structure) is replaced. System innovation tends to create a new system, for example, consumption is organised in such a way that the environment does not suffer as a result and other economic rules and principles are applied. 25 Dangelico and Pontrandolfo (2010:1610) conducted a review of research conducted over two decades on the features of green products. Their review is presented Table 2.2. Table 2.2: Studies on the characteristics of green products Authors Characteristics associated with the ‘green’ nature of a product Ljungberg (2007)  Reduce the materials and the use of energy for a product  Reduce emissions, dispersion and creation of toxins  Increase the amount of recyclable materials  Maximise the sustainable use of renewable resources  Minimise the service intensity for products and services  Extend the useful life of a product  Assess and minimize the environmental impact  Having a ‘functional economy’  Use ‘reverse logistics’  Increase the efficiency in the usage phase Luttropp and Lagerstedt (2006)  Do not use toxic substances and utilise closed loops for necessary but toxic ones  Minimise energy and resource consumption in the production phase and transport  Use structural features and high-quality materials to minimize weight  Minimise energy and resource consumption in the usage phase  Promote repair and upgrading  Promote long life  Invest in better materials, surface treatments or structural arrangements  Prearrange upgrading, repair and recycling  Promote upgrading, repair and recycling  Use as few joining elements as possible Roy, Wield, Gardiner and Potter (1996)  Capable of lessening global environmental problems  Energy efficient  Easily repairable  Designed to last, or to be reused, reconditioned or recycled  Generates minimum pollution and waste  Can be disposed of safely  Minimal use of materials, including packaging  Manufactured from renewable or abundant resources or recycled materials  Manufactured, if possible, locally and from locally obtainable materials to reduce transport requirements 26  Environmental information on product available to purchase  Not harmful to human health  Satisfies a genuine human need Shrivastava and Hart (1995)  Low environmental impact during usage  Easily composted, reused, or recycled at the end of their useful life Robert (1995)  Minimise the use of nonrenewable materials  Avoid the use of toxic materials  Use renewable resources in accordance with their rate of replenishment Peattie (1995)  Recyclability  Resource efficiency  Emissions  Impact on ecosystems  Social impact  Sustainability of resource use  Waste and disposal  Eco-efficiency of production and organisation Schmieding (1992)  Eliminate or replace product  Eliminate or reduce harmful ingredients  Substitute environmentally preferred materials or processes  Decrease weight or reduce volume  Produce concentrated product  Produce in bulk  Combine the functions of more than one product  Produce fewer models or styles  Redesign for more efficient use  Increase product life span  Reduce wasteful packaging  Improve reparability  Redesign for consumer reuse  Remanufacture the product Simon (1992)  Reduced raw material, high recycled content  Non-polluting manufacture/non-toxic materials  No unnecessary animal testing  No impact on protected species  Low energy consumption during production/use/disposal  Minimal or no packaging  Reuse//refillability where possible  Long useful life, updating capability  Post-consumer collection/disassembly system  Remanufacturing capability Elkington and Hailes (1988)  Not endangering the health of the consumer or of others  Causing no significant damage to the environment during manufacture use or disposal 27  Not consuming a disproportionate amount of energy during manufacture, use and disposal  Not causing unnecessary waste, either because of over-packaging or because of an unduly short useful life  No use of materials derived from threatened species or from threatened environments  Not involving unnecessary use or cruelty to animals  Not adversely affecting other countries, particularly the third world Source: Dangelico and Pontrandolfo (2010:1610) It is important to note that recent studies (American Marketing Association, 2013; Leonidou, Katsikeas and Morgan, 2013:152) show the same or similar characteristics as described by previous authors in Table 2.1. These characteristics indicate that irrespective of how green products have been classified, their contribution to the environment and their sustainability cannot be underestimated. 2.2.3 Benefits of Green Product Consumption Green products consumption is beneficial to the environment, the consumer, the economy and business. Environmentally, green products have evolved as a result of the growing concerns about global and local pollution levels, global warming, diminishing natural reserves and overflowing waste (Maniatis, 2015:218). Given these concerns, and according to the facts in Table 2.1, the consumption of green products will undoubtedly protect the environment. For example, green products have been proven to have reduced harmful side effects, reduced hazards, reduced toxic substances, reduced health issues, improved recyclability and improved environmental friendliness (Azevedo, Carvalho and Machado, 2011:851). Green standards, technologies, and practices in supply chains have benefitted consumers in many ways (Palevich, 2012:63). Over a period, economic gains could also be realised because of the benefits of green products consumption to consumers (Azevedo et al., 2011:851). For example, improved recyclability helps in reducing waste disposal costs (Azevedo et al., 2011:851). The true gains from a green product come from its lifecycle benefits (Zhao, Gao, Wu, Wang and Zhu, 2014:144). Many environmental impacts are caused by using natural resources in massive quantities, hazardous methods of manufacturing, harmful ways of usage, harmful patterns of generating waste and harmful patterns of disposal (Zhao et al., 2014: 143). A green product may be costlier than other comparative products but may have lower lifecycle 28 costs (Boztepe, 2012:92). For example, the product may be easily recyclable causing little negative impact on the environment (Zhao et al., 2014:143). Green products, their consumption and marketing also benefit business, especially in terms of providing competitive advantage. Prior studies have found that product and process innovations contribute to a firm’s competitive advantages (Chiou, Chan, Lettice and Chung 2011:822; Wong, 2012:469). These advantages, as postulated by Porter (1985), can be gained by differentiation and cost leadership strategies. ‘Greenness’ is a differentiator because waste and pollution represent an inefficient and incomplete use of resources. Thus, preventing or reducing waste and pollution through green marketing entails better resources utilisation, a lesser burden on the environment and better value to customers (Porter and van der Linde, 1995). While the innovative processes which are green confer cost advantage on a firm over its competitors, innovative green products also enhance firm profitability as products marketed under the green and innovative concept may bring in new customers and fresh revenue (Chiou et al., 2011, Kash and Rycoft, 2000; Lieberman and Montgomery, 1988; Nassimbeni, 2003). Studies have found that to embark on green innovation early enables a firm to gain first-mover advantage (Porter and van der Linde, 1995). For example, when Sony launched its ‘Green Partner’ programme in 2005 to restrict the use of hazardous substances by suppliers, only a few component manufacturers could meet this requirement. Those that met the requirement became Sony’s preferred suppliers and partners in developing the new green electronics market. This partnership enabled the chosen suppliers to gain first-mover advantages and to grow and develop with Sony in the flourishing green electronics arena. Many of the selected suppliers won orders from other green innovative companies such as Apple and Cisco by leveraging on the partnership status conferred on them by Sony. In the past ten years, the number of green electronic component manufacturers eligible to become Sony’s ‘Green Partners’ continues to expand. The late-movers face an uphill battle of competing against the first- and fast-movers. Competitive advantage is the result of matching the core competencies of a firm to market opportunities (Porter, 1985). Most firms, after identifying a market opportunity, take their first steps in building up green core competences by initiating green process innovation alongside product development. These steps may typically include activities such as promoting firm-wide learning about green innovation and environmental management trends and regulations, 29 translating green opportunities into green product features and building up new capabilities and aligning both new and existing capabilities with opportunities (Chen, 2008; Chiou et al., 2011). Green innovation usually entails changes in technology and raw materials and ultimately provides competitive advantage. For example, before the turn of the century, solder alloys with 40% lead were a primary choice for assembling electronics. Due to the relatively high toxicity of lead to human health and environment, European countries banned the use of lead in electronic assemblies in 2005. The leading component manufacturers responded by realigning their capabilities to facilitate a phased replacement of lead-based with lead-free solders in full compliance with the European requirements (United States Environmental Protection Agency, 2011). For manufacturers which were less capable of mastering the use of lead-free solder, it took them considerably longer to make the process and materials change to catch up with the market. Considering the benefits of green products to businesses, initiatives for green marketing are being forged. 2.2.4 Defining Green Marketing Green marketing is defined as holistic management for identifying, anticipating and satisfying the requirements of customers and society in a profitable and sustainable way (Choudhary and Gokarn, 2013:27; Stenzel 2011:1). According to Cheung, Lam and Lau (2015:234), green marketing refers to marketing activities developed to stimulate and maintain consumers’ environmental attitudes and behaviours. The American Marketing Association (AMA, 2015:2) provides three perspectives of green marketing, viewing it “as the marketing of products that are presumed to be environmentally safe” (retailing perspective), as “the development and marketing of products designed to minimise negative effects on the physical environment or to improve its quality” (social marketing perspective) and also as “the efforts by organisations to produce, promote, package, and reclaim products in a manner that is sensitive or responsive to ecological concerns” (environment perspective). Following the social marketing view, Morel and Kwakye (2012:53) state that that the primary objective of green marketing is to educate customers with an aim of increasing their willingness to go green, because this has an influence on changing the lifestyle and behaviour of potential consumers. Kotler and Armstrong (2013:453) posit that since marketing is seen as a process whereby the marketing mix (product, price, promotion and place) is used to respond to the needs and wants 30 of customers while achieving business objectives, many marketers have seen green marketing as simply another way of satisfying consumer needs under the same marketing mix. However, a closer look at the concept of green marketing shows a distinct variation that has transformed traditional marketing thinking (Kotler and Armstrong 2013:454). The term ‘green marketing’ itself has undergone many changes, with different terminology such as environmental marketing, sustainable marketing and ecological marketing being vaguely used. The term ‘green marketing’ surfaced in the 1980s following growi